From johannes at zellner.org  Tue Sep  1 00:43:50 2009
From: johannes at zellner.org (Dr. Johannes Zellner)
Date: Tue, 1 Sep 2009 07:43:50 +0200
Subject: grid line style
Message-ID: <103840bf0908312243n717802d7u1eaecf06e4adba0d@mail.gmail.com>

Hi,

how can I set the grid line style to a solid gray line? (octave 3.2 with
gnuplot)

-- 
Johannes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090901/92a608f2/attachment.html 

From jwe at octave.org  Tue Sep  1 01:02:37 2009
From: jwe at octave.org (John W. Eaton)
Date: Tue, 1 Sep 2009 02:02:37 -0400
Subject: grid line style
In-Reply-To: <103840bf0908312243n717802d7u1eaecf06e4adba0d@mail.gmail.com>
References: <103840bf0908312243n717802d7u1eaecf06e4adba0d@mail.gmail.com>
Message-ID: <19100.47357.934173.170780@segfault.lan>

On  1-Sep-2009, Dr. Johannes Zellner wrote:

| how can I set the grid line style to a solid gray line? (octave 3.2 with
| gnuplot)

Does gnuplot allow setting the line style for the grid lines?

jwe

From dbateman at dbateman.org  Tue Sep  1 01:07:58 2009
From: dbateman at dbateman.org (David Bateman)
Date: Tue, 01 Sep 2009 08:07:58 +0200
Subject: large sparse matrix
In-Reply-To: <67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>
	<25225293.post@talk.nabble.com>
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>
	<4A9C2A53.2060607@dbateman.org>
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
Message-ID: <4A9CBA3E.8060909@dbateman.org>

Carlo de Falco wrote:
>
> On 31 Aug 2009, at 21:53, David Bateman wrote:
>>
>> Yes, but the configuration of SuiteSparse in terms of the pivot 
>> threshold is the default that TIm Davis sets for Octave and whatever 
>> mathworks chose to use for matlab. So don't expect exactly the behavior.
> I see...
>
>> In Octave what does
>>
>> matrix_type(A)
>> give?
>
>
> >> matrix_type (A)
> ans = Positive Definite
> >>
>
>> If it doesn't flag your matrix as PD then there might be an issue 
>> with the detection of the matrix and PD and the solver used in Octave 
>> will be LU.
>
> on the other hand what surprises me is that:
>
> >> issparse (B)
> ans =  1
> >> matrix_type (B)
> ans = Rectangular
> >> issparse (A)
> ans =  1
> >> mat = [A B' spalloc(rows(A),1,0); B spalloc(rows(B),columns(A),1) 
> E'; spalloc(columns(A),1,1) E 0];
> >> matrix_type (mat)
> ans = Full
> >> issparse (mat)
> ans =  1
>
> why is mat marked as full???
>
>> Also, what does the code
>>
>> [R, p, q] = chol(A);
>> nnz(R)
>>
>
>> work for Octave and Matlab?
> I tried this in Octave, after 30 min it did not finish yet...
> I will only be able to try it on Matlab tomorrow.
>
>> If so what is the value of NNZ of the factorization in both cases.
>>
>> Regards
>> David
>
>
> Thanks,
> Carlo
>
What are the sizes of A, B and E?

D.


-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From johannes at zellner.org  Tue Sep  1 01:45:52 2009
From: johannes at zellner.org (Dr. Johannes Zellner)
Date: Tue, 1 Sep 2009 08:45:52 +0200
Subject: grid line style
In-Reply-To: <19100.47357.934173.170780@segfault.lan>
References: <103840bf0908312243n717802d7u1eaecf06e4adba0d@mail.gmail.com>
	<19100.47357.934173.170780@segfault.lan>
Message-ID: <103840bf0908312345m4ca8fc47kc3b0706f5762c490@mail.gmail.com>

gnuplot> set grid lt 1 lc rgb '#aaaaaa'

makes a solid gray grid.

-- 
Johannes


2009/9/1 John W. Eaton <jwe at octave.org>

> On  1-Sep-2009, Dr. Johannes Zellner wrote:
>
> | how can I set the grid line style to a solid gray line? (octave 3.2 with
> | gnuplot)
>
> Does gnuplot allow setting the line style for the grid lines?
>
> jwe
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090901/19c1c975/attachment.html 

From highegg at gmail.com  Tue Sep  1 02:14:02 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Tue, 1 Sep 2009 09:14:02 +0200
Subject: large sparse matrix
In-Reply-To: <67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>
	<25225293.post@talk.nabble.com>
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>
	<4A9C2A53.2060607@dbateman.org>
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
Message-ID: <69d8d540909010014n5ad9cad7k19d310149dc2b141@mail.gmail.com>

On Tue, Sep 1, 2009 at 1:12 AM, Carlo de Falco<carlo.defalco at gmail.com> wrote:
>
> On 31 Aug 2009, at 21:53, David Bateman wrote:
>>
>> Yes, but the configuration of SuiteSparse in terms of the pivot
>> threshold is the default that TIm Davis sets for Octave and whatever
>> mathworks chose to use for matlab. So don't expect exactly the
>> behavior.
> I see...
>
>> In Octave what does
>>
>> matrix_type(A)
>> give?
>
>
> ?>> matrix_type (A)
> ans = Positive Definite
> ?>>
>
>> If it doesn't flag your matrix as PD then there might be an issue
>> with the detection of the matrix and PD and the solver used in
>> Octave will be LU.
>
> on the other hand what surprises me is that:
>
> ?>> issparse (B)
> ans = ?1
> ?>> matrix_type (B)
> ans = Rectangular
> ?>> issparse (A)
> ans = ?1
> ?>> mat = [A B' spalloc(rows(A),1,0); B spalloc(rows(B),columns(A),1)
> E'; spalloc(columns(A),1,1) E 0];
> ?>> matrix_type (mat)
> ans = Full
> ?>> issparse (mat)
> ans = ?1
>
> why is mat marked as full???
>

Full is what you get when nothing special about the matrix is
discovered. Did you expect something else? The matrix is obviously not
positive definite, as seen from its construction, and I can see no
apparent other structure (triangularity or bandedness), though it of
course depends on the values of A, B and E.

hth

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From carlo.defalco at gmail.com  Tue Sep  1 02:57:37 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Tue, 1 Sep 2009 09:57:37 +0200
Subject: large sparse matrix
In-Reply-To: <69d8d540909010014n5ad9cad7k19d310149dc2b141@mail.gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>
	<25225293.post@talk.nabble.com>
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>
	<4A9C2A53.2060607@dbateman.org>
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
	<69d8d540909010014n5ad9cad7k19d310149dc2b141@mail.gmail.com>
Message-ID: <44BC9295-727A-467B-956E-1FE4CD2F7F2B@gmail.com>


On 1 Sep 2009, at 09:14, Jaroslav Hajek wrote:

> On Tue, Sep 1, 2009 at 1:12 AM, Carlo de  
> Falco<carlo.defalco at gmail.com> wrote:
>>
>> On 31 Aug 2009, at 21:53, David Bateman wrote:
>>>
>>> Yes, but the configuration of SuiteSparse in terms of the pivot
>>> threshold is the default that TIm Davis sets for Octave and whatever
>>> mathworks chose to use for matlab. So don't expect exactly the
>>> behavior.
>> I see...
>>
>>> In Octave what does
>>>
>>> matrix_type(A)
>>> give?
>>
>>
>>  >> matrix_type (A)
>> ans = Positive Definite
>>  >>
>>
>>> If it doesn't flag your matrix as PD then there might be an issue
>>> with the detection of the matrix and PD and the solver used in
>>> Octave will be LU.
>>
>> on the other hand what surprises me is that:
>>
>>  >> issparse (B)
>> ans =  1
>>  >> matrix_type (B)
>> ans = Rectangular
>>  >> issparse (A)
>> ans =  1
>>  >> mat = [A B' spalloc(rows(A),1,0); B spalloc(rows(B),columns(A),1)
>> E'; spalloc(columns(A),1,1) E 0];
>>  >> matrix_type (mat)
>> ans = Full
>>  >> issparse (mat)
>> ans =  1
>>
>> why is mat marked as full???
>>
>
> Full is what you get when nothing special about the matrix is
> discovered. Did you expect something else? The matrix is obviously not
> positive definite, as seen from its construction, and I can see no
> apparent other structure (triangularity or bandedness), though it of
> course depends on the values of A, B and E.
>
> hth

Well, the matrix is of course symmetric by construction,
but I don't think there's much more...
But actually the reason I was surprised that it is marked as
"Full" is that I thought this mark had to do with its storage format.
Maybe another name like, e.g., "Generic" would have been less  
misleading...
Thnaks for the info!
c.







From carlo.defalco at gmail.com  Tue Sep  1 03:04:17 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Tue, 1 Sep 2009 10:04:17 +0200
Subject: large sparse matrix
In-Reply-To: <67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>
	<25225293.post@talk.nabble.com>
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>
	<4A9C2A53.2060607@dbateman.org>
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
Message-ID: <420ba0760909010104x3c7385fdi39eab8d17440a06f@mail.gmail.com>

2009/9/1 Carlo de Falco <carlo.defalco at gmail.com>:

>> Also, what does the code
>>
>> [R, p, q] = chol(A);
>> nnz(R)
>>
>
>> work for Octave and Matlab?
>
> I tried this in Octave, after 30 min it did not finish yet...
> I will only be able to try it on Matlab tomorrow.
>
>> If so what is the value of NNZ of the factorization in both cases.

I killed the Octave after it had been running more than one hour and
produced no result.
In Matlab it ran in less than a second and the result was:

>> [R,p,q]=chol(A);
>> nnz (R)
ans =
     1368019
>> issparse (R)
ans =
     1
>> nnz(R)/prod(size(A))
ans =
     7.035863106552923e-03

Thanks,
c.

From carlo.defalco at gmail.com  Tue Sep  1 03:12:12 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Tue, 1 Sep 2009 10:12:12 +0200
Subject: large sparse matrix
In-Reply-To: <420ba0760909010104x3c7385fdi39eab8d17440a06f@mail.gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>
	<25225293.post@talk.nabble.com>
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>
	<4A9C2A53.2060607@dbateman.org>
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
	<420ba0760909010104x3c7385fdi39eab8d17440a06f@mail.gmail.com>
Message-ID: <B573A6D7-B3E4-425C-9DEC-6E1F2188F233@gmail.com>


On 1 Sep 2009, at 10:04, Carlo de Falco wrote:

> 2009/9/1 Carlo de Falco <carlo.defalco at gmail.com>:
>
>>> Also, what does the code
>>>
>>> [R, p, q] = chol(A);
>>> nnz(R)
>>>
>>
>>> work for Octave and Matlab?
>>
>> I tried this in Octave, after 30 min it did not finish yet...
>> I will only be able to try it on Matlab tomorrow.
>>
>>> If so what is the value of NNZ of the factorization in both cases.
>
> I killed the Octave after it had been running more than one hour and
> produced no result.
> In Matlab it ran in less than a second and the result was:
>
>>> [R,p,q]=chol(A);
>>> nnz (R)
> ans =
>     1368019
>>> issparse (R)
> ans =
>     1
>>> nnz(R)/prod(size(A))
> ans =
>     7.035863106552923e-03

oops, what I actually meant to show here was

 >> nnz(R)/prod(size(R))
ans=
    1.407374482251814e-02

> Thanks,
c.


From dbateman at dbateman.org  Tue Sep  1 14:09:04 2009
From: dbateman at dbateman.org (David Bateman)
Date: Tue, 01 Sep 2009 21:09:04 +0200
Subject: large sparse matrix
In-Reply-To: <420ba0760909010104x3c7385fdi39eab8d17440a06f@mail.gmail.com>
References: <420ba0760908310753i7fe60591l9d4d657a6db9dade@mail.gmail.com>	
	<25225293.post@talk.nabble.com>	
	<D99E8142-388B-42AC-8DE2-0CB23F271C82@gmail.com>	
	<4A9C2A53.2060607@dbateman.org>	
	<67814968-724A-4C38-89A7-01E2D4613C6D@gmail.com>
	<420ba0760909010104x3c7385fdi39eab8d17440a06f@mail.gmail.com>
Message-ID: <4A9D7150.4000000@dbateman.org>

Carlo de Falco wrote:
> 2009/9/1 Carlo de Falco <carlo.defalco at gmail.com>:
>
>   
>>> Also, what does the code
>>>
>>> [R, p, q] = chol(A);
>>> nnz(R)
>>>
>>>       
>>> work for Octave and Matlab?
>>>       
>> I tried this in Octave, after 30 min it did not finish yet...
>> I will only be able to try it on Matlab tomorrow.
>>
>>     
>>> If so what is the value of NNZ of the factorization in both cases.
>>>       
>
> I killed the Octave after it had been running more than one hour and
> produced no result.
> In Matlab it ran in less than a second and the result was:
>
>   
>>> [R,p,q]=chol(A);
>>> nnz (R)
>>>       
> ans =
>      1368019
>   
>>> issparse (R)
>>>       
> ans =
>      1
>   
>>> nnz(R)/prod(size(A))
>>>       
> ans =
>      7.035863106552923e-03
>
> Thanks,
> c.
>
>   
Could you sent me the code to generate A,B,E and mat... I might have a 
better chance of figuring out the issue in that case..

Regards
davis


-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From jwe at octave.org  Tue Sep  1 14:58:44 2009
From: jwe at octave.org (John W. Eaton)
Date: Tue, 1 Sep 2009 15:58:44 -0400
Subject: grid line style
In-Reply-To: <103840bf0908312345m4ca8fc47kc3b0706f5762c490@mail.gmail.com>
References: <103840bf0908312243n717802d7u1eaecf06e4adba0d@mail.gmail.com>
	<19100.47357.934173.170780@segfault.lan>
	<103840bf0908312345m4ca8fc47kc3b0706f5762c490@mail.gmail.com>
Message-ID: <19101.31988.387738.731001@segfault.lan>

On  1-Sep-2009, Dr. Johannes Zellner wrote:

| gnuplot> set grid lt 1 lc rgb '#aaaaaa'
| 
| makes a solid gray grid.

OK, so it would be possible with the gnuplot backend.  But what
properties should affect the linestyle for the grid lines?  I don't
think Matlab has any that are independent of the line styles of the
axes lines themselves.  So we would need to come up with something,
which is likelyt to introduce compatibility problems in the future.
If you'd like to discuss the development of a feature like this in
Octave, then plase start a thread on the maintainers list.

Of course a workaround is to write your data to a file and use gnuplt
directly...

jwe

From jthoo at yccd.edu  Tue Sep  1 16:11:10 2009
From: jthoo at yccd.edu (John B. Thoo)
Date: Tue, 1 Sep 2009 14:11:10 -0700
Subject: ode45 gives solution in the "Dark Side," but not in Octave
Message-ID: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>

Hi.  I have a question about using ode45 in Octave 3.2.2 (and OdePkg  
on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a  
server ... Linux?).  My code can be found here at the moment:

<http://ms.yccd.edu/~jb2/PickUp/twoway.zip>

"run_twoway.m" executes the code.

Both Octave and the Dark Side solve the problem (i.e., do not return  
an error) to produce a solution  u,  and both give

 > size (u)
ans =

    8192    102

However, while I can plot the solution in the Dark Side, in Octave I  
can plot only the first four frames.  In fact, in Octave I get  NaN +  
0.00000i  from  u(:,5)  on.  Why is that?

I would appreciate any hints.  Thanks.

---John.

From jwe at octave.org  Tue Sep  1 17:00:18 2009
From: jwe at octave.org (John W. Eaton)
Date: Tue, 1 Sep 2009 18:00:18 -0400
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
Message-ID: <19101.39282.859417.679486@segfault.lan>

On  1-Sep-2009, John B. Thoo wrote:

| Hi.  I have a question about using ode45 in Octave 3.2.2 (and OdePkg  
| on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a  
| server ... Linux?).  My code can be found here at the moment:
| 
| <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
| 
| "run_twoway.m" executes the code.
| 
| Both Octave and the Dark Side solve the problem (i.e., do not return  
| an error) to produce a solution  u,  and both give
| 
|  > size (u)
| ans =
| 
|     8192    102
| 
| However, while I can plot the solution in the Dark Side, in Octave I  
| can plot only the first four frames.  In fact, in Octave I get  NaN +  
| 0.00000i  from  u(:,5)  on.  Why is that?
| 
| I would appreciate any hints.  Thanks.

Have you tried using lsode in Octave instead of ode45 (which is not
part of Octave, BTW).

jwe

From Mike.Coyne at PACCAR.com  Wed Sep  2 10:54:30 2009
From: Mike.Coyne at PACCAR.com (Mike Coyne)
Date: Wed, 2 Sep 2009 10:54:30 -0500
Subject: FW: mex files built with fortran having problmes linking in
 3.2.2
Message-ID: <A869E1B24D244D4CA381EFFE394296E8071F38E5@PBCDEPMXM5.na.paccar.com>


From: Mike Coyne 
Sent: Tuesday, September 01, 2009 11:31 AM
To: help-octave at octave.org
Subject: mex files built with fortran having problmes linking in 3.2.2

I am having problems createing a  "mex" file that can access the mx
functions .

A simple example is in the dist examples
 mkoctfile --mex myfevalf.f

octave --quiet --eval myfevalf
warning: mark_as_command is obsolete and will be removed from a future
version of Octave
error:
/afs/pbcell/home/mcoyne/.gnome-desktop/octave-3.2.2/octave-3.2.2/example
s/myfevalf.mex: undefined symbol: mxgetstring_
terminate called after throwing an instance of
'octave_execution_exception'
panic: Aborted -- stopping myself...
attempting to save variables to `octave-core'...
save to `octave-core' complete
Aborted

the mxgetstring_ is the fortran call to the c++  mxGetString ? 
How do i build ? is there a library to link against? Or alteritibly what
would be the correct wrapper function to generate the mxgetstring_
function calling the C++ mxGetString function?

My platform is Rhel5 w/ gnu gcc/gfortran  4.4.1 



From Thomas.Treichl at gmx.net  Wed Sep  2 13:28:54 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Wed, 02 Sep 2009 20:28:54 +0200
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
Message-ID: <4A9EB966.7050203@gmx.net>

John B. Thoo schrieb:
> Hi.  I have a question about using ode45 in Octave 3.2.2 (and OdePkg  
> on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a  
> server ... Linux?).  My code can be found here at the moment:
> 
> <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
> 
> "run_twoway.m" executes the code.
> 
> Both Octave and the Dark Side solve the problem (i.e., do not return  
> an error) to produce a solution  u,  and both give
> 
>  > size (u)
> ans =
> 
>     8192    102
> 
> However, while I can plot the solution in the Dark Side, in Octave I  
> can plot only the first four frames.  In fact, in Octave I get  NaN +  
> 0.00000i  from  u(:,5)  on.  Why is that?
> 
> I would appreciate any hints.  Thanks.
> 
> ---John.

Good question, I would say that you have a tricky problem and the solver gets 
into trouble ;) For a quick workaround: what happens if you use another solver 
like ode78? What about ode5r? I tried your example but get

   octave-3.2.0:10> run_twoway
   error: `heaviside' undefined near line 4 column 18
   error: called from:
   error:   /Users/Thomas/tmp/thoo/TwoWay/make_k.m at line 4, column 14
   error:   /Users/Thomas/tmp/thoo/TwoWay/run_twoway.m at line 20, column 3

Best regards

   Thomas

From jwe at octave.org  Wed Sep  2 14:00:13 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 2 Sep 2009 15:00:13 -0400
Subject: FW: mex files built with fortran having problmes linking in
 3.2.2
In-Reply-To: <A869E1B24D244D4CA381EFFE394296E8071F38E5@PBCDEPMXM5.na.paccar.com>
References: <A869E1B24D244D4CA381EFFE394296E8071F38E5@PBCDEPMXM5.na.paccar.com>
Message-ID: <19102.49342.644.703561@segfault.lan>

On  2-Sep-2009, Mike Coyne wrote:

| From: Mike Coyne 
| Sent: Tuesday, September 01, 2009 11:31 AM
| To: help-octave at octave.org
| Subject: mex files built with fortran having problmes linking in 3.2.2
| 
| I am having problems createing a  "mex" file that can access the mx
| functions .
| 
| A simple example is in the dist examples
|  mkoctfile --mex myfevalf.f
| 
| octave --quiet --eval myfevalf
| warning: mark_as_command is obsolete and will be removed from a future
| version of Octave
| error:
| /afs/pbcell/home/mcoyne/.gnome-desktop/octave-3.2.2/octave-3.2.2/example
| s/myfevalf.mex: undefined symbol: mxgetstring_
| terminate called after throwing an instance of
| 'octave_execution_exception'
| panic: Aborted -- stopping myself...
| attempting to save variables to `octave-core'...
| save to `octave-core' complete
| Aborted
| 
| the mxgetstring_ is the fortran call to the c++  mxGetString ? 
| How do i build ? is there a library to link against? Or alteritibly what
| would be the correct wrapper function to generate the mxgetstring_
| function calling the C++ mxGetString function?
| 
| My platform is Rhel5 w/ gnu gcc/gfortran  4.4.1 

Octave doesn't provide Fortran bindings for the MEX interface.  I
consider this a missing feature, not a bug.  I would of course welcome
a patch to add them.  But before submitting a patch, it would probably
be best to discuss the implementation on the maintainers list.

The example files are in the distribution because there was some (very
incomplete) support for Fortran MEX files at one time, but it was
removed when the C MEX interface was finished.  I guess we forgot to
remove the Fortran example files and/or make a note in the docs.

jwe

From dimitry at latinmail.com  Wed Sep  2 14:25:56 2009
From: dimitry at latinmail.com (dimitry A.)
Date: Wed, 02 Sep 2009 21:25:56 +0200
Subject: No subject
Message-ID: <200909021926.n82JQ37M014807@mx1.cae.wisc.edu>

An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090902/af5e844a/attachment.html 

From jthoo at yccd.edu  Wed Sep  2 18:55:40 2009
From: jthoo at yccd.edu (John B. Thoo)
Date: Wed, 2 Sep 2009 16:55:40 -0700
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <19101.39282.859417.679486@segfault.lan>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<19101.39282.859417.679486@segfault.lan>
Message-ID: <25EC71B0-6023-43E8-8A95-5488C5CEB9E6@yccd.edu>


On Sep 1, 2009, at 3:00 PM, John W. Eaton wrote:

> On  1-Sep-2009, John B. Thoo wrote:
>
> | Hi.  I have a question about using ode45 in Octave 3.2.2 (and OdePkg
> | on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a
> | server ... Linux?).  My code can be found here at the moment:
> |
> | <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
> |
> | "run_twoway.m" executes the code.
> |
> | Both Octave and the Dark Side solve the problem (i.e., do not return
> | an error) to produce a solution  u,  and both give
> |
> |  > size (u)
> | ans =
> |
> |     8192    102
> |
> | However, while I can plot the solution in the Dark Side, in Octave I
> | can plot only the first four frames.  In fact, in Octave I get   
> NaN +
> | 0.00000i  from  u(:,5)  on.  Why is that?
> |
> | I would appreciate any hints.  Thanks.
>
> Have you tried using lsode in Octave instead of ode45 (which is not
> part of Octave, BTW).

I haven't tried lsode.  The reason for my using ode45 is my advisor  
uses the Dark Side only, and we need to run the same code.  Thanks.

---John.

From jthoo at yccd.edu  Wed Sep  2 18:58:14 2009
From: jthoo at yccd.edu (John B. Thoo)
Date: Wed, 2 Sep 2009 16:58:14 -0700
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <4A9EB966.7050203@gmx.net>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<4A9EB966.7050203@gmx.net>
Message-ID: <19FD46E0-2D97-4FA3-BAA6-2938D34AB6A0@yccd.edu>


On Sep 2, 2009, at 11:28 AM, Thomas Treichl wrote:

> John B. Thoo schrieb:
>> Hi.  I have a question about using ode45 in Octave 3.2.2 (and  
>> OdePkg  on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a)  
>> (on a  server ... Linux?).  My code can be found here at the moment:
>> <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
>> "run_twoway.m" executes the code.
>> Both Octave and the Dark Side solve the problem (i.e., do not  
>> return  an error) to produce a solution  u,  and both give
>>  > size (u)
>> ans =
>>     8192    102
>> However, while I can plot the solution in the Dark Side, in Octave  
>> I  can plot only the first four frames.  In fact, in Octave I get   
>> NaN +  0.00000i  from  u(:,5)  on.  Why is that?
>> I would appreciate any hints.  Thanks.
>> ---John.
>
> Good question, I would say that you have a tricky problem and the  
> solver gets into trouble ;) For a quick workaround: what happens if  
> you use another solver like ode78? What about ode5r? I tried your  
> example but get
>
>   octave-3.2.0:10> run_twoway
>   error: `heaviside' undefined near line 4 column 18
>   error: called from:
>   error:   /Users/Thomas/tmp/thoo/TwoWay/make_k.m at line 4, column 14
>   error:   /Users/Thomas/tmp/thoo/TwoWay/run_twoway.m at line 20,  
> column 3
>
> Best regards
>
>   Thomas

Hi, Thomas.  The function "heaviside" is part of SpecFun package:

<http://octave.sourceforge.net/doc/funref_specfun.html>

I will try the other solvers you suggest.  Thanks.

---John.

From jwe at octave.org  Wed Sep  2 19:30:34 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 2 Sep 2009 20:30:34 -0400
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <25EC71B0-6023-43E8-8A95-5488C5CEB9E6@yccd.edu>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<19101.39282.859417.679486@segfault.lan>
	<25EC71B0-6023-43E8-8A95-5488C5CEB9E6@yccd.edu>
Message-ID: <19103.3626.250351.795124@segfault.lan>

On  2-Sep-2009, John B. Thoo wrote:

| On Sep 1, 2009, at 3:00 PM, John W. Eaton wrote:
| 
| > On  1-Sep-2009, John B. Thoo wrote:
| >
| > | Hi.  I have a question about using ode45 in Octave 3.2.2 (and OdePkg
| > | on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a
| > | server ... Linux?).  My code can be found here at the moment:
| > |
| > | <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
| > |
| > | "run_twoway.m" executes the code.
| > |
| > | Both Octave and the Dark Side solve the problem (i.e., do not return
| > | an error) to produce a solution  u,  and both give
| > |
| > |  > size (u)
| > | ans =
| > |
| > |     8192    102
| > |
| > | However, while I can plot the solution in the Dark Side, in Octave I
| > | can plot only the first four frames.  In fact, in Octave I get   
| > NaN +
| > | 0.00000i  from  u(:,5)  on.  Why is that?
| > |
| > | I would appreciate any hints.  Thanks.
| >
| > Have you tried using lsode in Octave instead of ode45 (which is not
| > part of Octave, BTW).
| 
| I haven't tried lsode.  The reason for my using ode45 is my advisor  
| uses the Dark Side only, and we need to run the same code.  Thanks.

Here's a simple wrapper for lsode that I think will provide a simple
ode45 interface.  It doesn't cover all the options, but it will
probably work in a lot of cases, provided that you just need a
solution, and not specifically a Runge-Kutta method.  It shoudl be
fairly easy to fix the interface if ti isn't compatible enough for
you.

jwe

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ode45.m
Url: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090902/6a2c382d/attachment.ksh 

From jthoo at yccd.edu  Thu Sep  3 07:18:35 2009
From: jthoo at yccd.edu (John B. Thoo)
Date: Thu, 3 Sep 2009 05:18:35 -0700
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <4A9EB966.7050203@gmx.net>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<4A9EB966.7050203@gmx.net>
Message-ID: <53FECAC2-F628-4A22-BC27-0ED9184EBCBB@yccd.edu>


On Sep 2, 2009, at 11:28 AM, Thomas Treichl wrote:

> John B. Thoo schrieb:
>> Hi.  I have a question about using ode45 in Octave 3.2.2 (and  
>> OdePkg  on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a)  
>> (on a  server ... Linux?).  My code can be found here at the moment:
>> <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
>> "run_twoway.m" executes the code.
>> Both Octave and the Dark Side solve the problem (i.e., do not  
>> return  an error) to produce a solution  u,  and both give
>>  > size (u)
>> ans =
>>     8192    102
>> However, while I can plot the solution in the Dark Side, in Octave  
>> I  can plot only the first four frames.  In fact, in Octave I get   
>> NaN +  0.00000i  from  u(:,5)  on.  Why is that?
>> I would appreciate any hints.  Thanks.
>> ---John.
>
> Good question, I would say that you have a tricky problem and the  
> solver gets into trouble ;) For a quick workaround: what happens if  
> you use another solver like ode78? What about ode5r? I tried your  
> example but get
>
>   octave-3.2.0:10> run_twoway
>   error: `heaviside' undefined near line 4 column 18
>   error: called from:
>   error:   /Users/Thomas/tmp/thoo/TwoWay/make_k.m at line 4, column 14
>   error:   /Users/Thomas/tmp/thoo/TwoWay/run_twoway.m at line 20,  
> column 3
>
> Best regards
>
>   Thomas

Hi, Thomas.

With ode78, I can plot up to three frames.  ode5r gives an error:

octave-3.2.2:11> run_twoway
error: `ode5r' undefined near line 36 column 11
error: called from:
error:   /Users/jbthoo/Desktop/NumMethPractice/MHD/TwoWay/ 
run_twoway.m at line 36, column 9
octave-3.2.2:11>

But you've given me an idea: I look in ~/octave/odepkg-0.6.7/, and  
see I also have ode23 and ode54.  Unfortunately,

With ode23, I can plot up to seven frames.
With ode54, I can plot up to four frames.

Thanks.

---John.

From jwe at octave.org  Thu Sep  3 07:21:16 2009
From: jwe at octave.org (John W. Eaton)
Date: Thu, 3 Sep 2009 08:21:16 -0400
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <53FECAC2-F628-4A22-BC27-0ED9184EBCBB@yccd.edu>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<4A9EB966.7050203@gmx.net>
	<53FECAC2-F628-4A22-BC27-0ED9184EBCBB@yccd.edu>
Message-ID: <19103.46268.500179.796623@segfault.lan>

On  3-Sep-2009, John B. Thoo wrote:

| With ode78, I can plot up to three frames.  ode5r gives an error:
| 
| octave-3.2.2:11> run_twoway
| error: `ode5r' undefined near line 36 column 11
| error: called from:
| error:   /Users/jbthoo/Desktop/NumMethPractice/MHD/TwoWay/ 
| run_twoway.m at line 36, column 9
| octave-3.2.2:11>
| 
| But you've given me an idea: I look in ~/octave/odepkg-0.6.7/, and  
| see I also have ode23 and ode54.  Unfortunately,
| 
| With ode23, I can plot up to seven frames.
| With ode54, I can plot up to four frames.

Did you try my wrapper for lsode?  I'm curious to know whether it can
solve your problem.

jwe

From Thomas.Treichl at gmx.net  Thu Sep  3 13:15:26 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Thu, 03 Sep 2009 20:15:26 +0200
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <19FD46E0-2D97-4FA3-BAA6-2938D34AB6A0@yccd.edu>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<4A9EB966.7050203@gmx.net>
	<19FD46E0-2D97-4FA3-BAA6-2938D34AB6A0@yccd.edu>
Message-ID: <4AA007BE.3010200@gmx.net>

John B. Thoo schrieb:
> Hi, Thomas.  The function "heaviside" is part of SpecFun package:
> 
> <http://octave.sourceforge.net/doc/funref_specfun.html>
> 
> I will try the other solvers you suggest.  Thanks.
> 
> ---John.

Ok, I was able to start your script - thanks for the hint. You're using fixed 
time steps (tspan) with the ode45 solver which is absolutely okay, but the next 
argument you use (w0) already is complex even before ode45 can do its first 
step, ie. the initial values already are complex. I don't expect that these 
values should be complex?! Here is what I see if I remove the semicolon at the 
end of the line w0=...

octave-3.2.0:1> run_twoway
   w0 =
     -0.537882842739990 + 0.000000000000000i
     -0.537882789290712 + 0.000000000000000i
     -0.537882628942835 + 0.000000000000000i
     -0.537882361696230 + 0.000000000000000i
     -0.537881987550684 - 0.000000000000000i
     -0.537881506505897 - 0.000000000000000i
     -0.537880918561486 + 0.000000000000000i
   <SNIP>

I think that this is the problem you have - ode45 thinks your problem is complex 
and then tries to solve your complex problem, which it is able to if the complex 
problem makes sense... So the origin of this problem seems to be somewhere else 
in your initialization routines. Can you confirm?

Best regards
   Thomas

PS. Just because of interest why you don't have ode5r, did you install OdePkg 
with the 'pkg' command or did you just unpack the *.tar.gz into some directory? 
ode5r should be there if you installed OdePkg (it is globally installed on my 
system but it can also be somewhere in your local home directory)

octave-3.1.55:7> which ode5r
`ode5r' is a function from the file 
/Users/Thomas/bin/Octave.app.3.2.0/Octave.app/Contents/Resources/libexec/octave/packages/odepkg-0.6.7/i386-apple-darwin8.11.1-api-v37/dldsolver.oct

From jthoo at yccd.edu  Thu Sep  3 13:33:48 2009
From: jthoo at yccd.edu (John B. Thoo)
Date: Thu, 3 Sep 2009 11:33:48 -0700
Subject: ode45 gives solution in the "Dark Side," but not in Octave
In-Reply-To: <19103.3626.250351.795124@segfault.lan>
References: <5674DAF1-8AED-4C51-A3BA-345713B9CA49@yccd.edu>
	<19101.39282.859417.679486@segfault.lan>
	<25EC71B0-6023-43E8-8A95-5488C5CEB9E6@yccd.edu>
	<19103.3626.250351.795124@segfault.lan>
Message-ID: <291B5407-2F95-4C63-AEC5-6C0129951092@yccd.edu>


On Sep 2, 2009, at 5:30 PM, John W. Eaton wrote:

> On  2-Sep-2009, John B. Thoo wrote:
>
> | On Sep 1, 2009, at 3:00 PM, John W. Eaton wrote:
> |
> | > On  1-Sep-2009, John B. Thoo wrote:
> | >
> | > | Hi.  I have a question about using ode45 in Octave 3.2.2 (and  
> OdePkg
> | > | on Mac OS X 10.4.11) vs. in the "Dark Side" 7.4.0 (R2007a) (on a
> | > | server ... Linux?).  My code can be found here at the moment:
> | > |
> | > | <http://ms.yccd.edu/~jb2/PickUp/twoway.zip>
> | > |
> | > | "run_twoway.m" executes the code.
> | > |
> | > | Both Octave and the Dark Side solve the problem (i.e., do not  
> return
> | > | an error) to produce a solution  u,  and both give
> | > |
> | > |  > size (u)
> | > | ans =
> | > |
> | > |     8192    102
> | > |
> | > | However, while I can plot the solution in the Dark Side, in  
> Octave I
> | > | can plot only the first four frames.  In fact, in Octave I get
> | > NaN +
> | > | 0.00000i  from  u(:,5)  on.  Why is that?
> | > |
> | > | I would appreciate any hints.  Thanks.
> | >
> | > Have you tried using lsode in Octave instead of ode45 (which is  
> not
> | > part of Octave, BTW).
> |
> | I haven't tried lsode.  The reason for my using ode45 is my advisor
> | uses the Dark Side only, and we need to run the same code.  Thanks.
>
> Here's a simple wrapper for lsode that I think will provide a simple
> ode45 interface.  It doesn't cover all the options, but it will
> probably work in a lot of cases, provided that you just need a
> solution, and not specifically a Runge-Kutta method.  It shoudl be
> fairly easy to fix the interface if ti isn't compatible enough for
> you.
>
> jwe
>
> function [t_out, x] = ode45 (f, t, x0, options)
>
>   if (nargin == 3 || nargin == 4)
>
>     global __ode45s_rhs_fcn__;
>     __ode45s_rhs_fcn__ = f;
>
>     if (nargin == 4)
>       if (isfield (options, "RelTol") && ! isempty (options.RelTol))
> 	lsode_options ("relative tolerance", options.RelTol)
>       endif
>       if (isfield (options, "AbsTol") && ! isempty (options.AbsTol))
> 	lsode_options ("absolute tolerance", options.AbsTol)
>       endif
>     endif
>
>     x = lsode (@__lsode_rhs__, x0, t);
>     t_out = t;
>
>   else
>     print_usage ();
>   endif
>
> endfunction
>
> %  
> ----------------------------------------------------------------------
>
> function xdot = __lsode_rhs__ (x, t)
>
>   global __ode45s_rhs_fcn__;
>
>   xdot = __ode45s_rhs_fcn__ (t, x);
>
> endfunction

Hi, jwe.  Thanks for taking time on this.

I'm sorry, but I'm not familiar with using a "wrapper," so I haven't  
tried it, yet.  Do I save the above to a .m file?  Then how do I call  
on it?

In any case, I tried lsode directly in my code, in which, making a  
guess for how to proceed, I replaced the lines in "run_twoway.m"

options = odeset ();
[t, w] = ode45 (@F, tspan, w0, options, k, lx, ltrunc, epsilon, mu);

with the line

[w, istate, msg] = lsode (@F, tspan, w0, k, lx, ltrunc, epsilon, mu);

But when I executed the code, I got this:

octave-3.2.2:5> run_twoway
error: Invalid call to lsode.  Correct usage is:

  -- Loadable Function: [X, ISTATE, MSG] = lsode (FCN, X_0, T, T_CRIT)


Additional help for built-in functions and operators is
available in the on-line version of the manual.  Use the command
`doc <topic>' to search the manual index.

Help and information about Octave is also available on the WWW
at http://www.octave.org and via the help at octave.org
mailing list.
octave-3.2.2:5>

Thanks again.

---John.

From dasergatskov at gmail.com  Thu Sep  3 17:09:05 2009
From: dasergatskov at gmail.com (Dmitri A. Sergatskov)
Date: Thu, 3 Sep 2009 17:09:05 -0500
Subject: [OctDev] test_sparse.m fails on ppc64 build
In-Reply-To: <loom.20090903T222522-569@post.gmane.org>
References: <loom.20090903T222522-569@post.gmane.org>
Message-ID: <892b16670909031509w39a3c627w158438ec186aba0b@mail.gmail.com>

On Thu, Sep 3, 2009 at 3:27 PM, Orion Poplawski<orion at cora.nwra.com> wrote:
> Running make check on octave 3.2.2 build in Fedora Rawhide fails in
> test_sparse.m on ppc64.
>
> Fixed test scripts:
> ?test_args.m ............................................ PASS ? 26/26
> .....
> ?test_sparse.m ..........................................panic: Segmentation
> fault -- stopping myself...
> make[2]: *** [check] Segmentation fault
>
>
> Fedora bug: https://bugzilla.redhat.com/show_bug.cgi?id=521142
> Build log: http://koji.fedoraproject.org/koji/getfile?taskID=1653016&name=build.log
>
> Any ideas? ?Any kind of test verbosity that can be turned on to help debug?
>
> Thanks!
>
> - Orion Poplawski
>

First of all i think it should go to octave mailing list rather than
to octave-forge
(cc accordingly).

Second, the lapack/atlas libraries on Fedora 11 and rawhide are suspect
(they are definitely broken on i386). Sometime ago somebody decided that
some files from lapack that should be compiled without optimization
(per lapack's docs) can be compiled with Os. This used to work for a while
but now gcc got more aggressive and breaks those on some platforms
(at least on i386 and may be on ppc).

Sincerely,

Dmitri.
--


From asmund.skjaveland at bio.uio.no  Fri Sep  4 06:40:25 2009
From: asmund.skjaveland at bio.uio.no (=?ISO-8859-1?Q?=C5smund_Skj=E6veland?=)
Date: Fri, 04 Sep 2009 13:40:25 +0200
Subject: plotting NaN values as blank in pcolor plots
Message-ID: <4AA0FCA9.4040102@ulrik.uio.no>

When I plot an array using pcolor, the NaN fields are plotted as the
lowest color in the colormap. How do I make them plot blank, like in matlab?

Versions:

gnuplot 4.2 patchlevel 4 (stock Ubuntu 9.04 package)
GNU Octave, version 3.2.2 (self-compiled)



-- 
?smund Skj?veland

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090904/bbe7167e/attachment.bin 

From marcio.alves.diniz at gmail.com  Fri Sep  4 06:29:12 2009
From: marcio.alves.diniz at gmail.com (=?ISO-8859-1?Q?M=E1rcio_Diniz?=)
Date: Fri, 4 Sep 2009 08:29:12 -0300
Subject: bfgsmin works for a version but not for the other
Message-ID: <933ae1930909040429k72bae91eka94f6dac1765c7b9@mail.gmail.com>

Hi there!
I am using Octave 2.9.9 and the bfgsmin works well. However, the same code
does not work when I try to run it in Octave 3.0.1 - the error message is
"warning: bfgsmin: failure, exiting. Try different start values?" even
though I am using the same starting values.
Is there any difference between the bfgsmin versions? If not, what is the
problem?
Thanks in advance
Marcio Diniz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090904/3314840f/attachment.html 

From dennis.eder at etas.com  Fri Sep  4 13:57:28 2009
From: dennis.eder at etas.com (dennis.eder at etas.com)
Date: Fri, 4 Sep 2009 20:57:28 +0200
Subject: =?ISO-8859-1?Q?AUTO=3A_Dennis_Eder_ist_au=DFer_Haus_=28R=FCckkehr_am?=
	=?ISO-8859-1?Q?_08=2E09=2E2009=29?=
Message-ID: <OFAB1C29E6.E6757E51-ONC1257627.00682396-C1257627.00682396@etas.de>


Ich bin bis 08.09.2009 abwesend




Hinweis: Dies ist eine automatische Antwort auf Ihre Nachricht  "
Help-octave Digest, Vol 42, Issue 5" gesendet am 04.09.2009 20:52:46.

Diese ist die einzige Benachrichtigung, die Sie empfangen werden, w?hrend
diese Person abwesend ist.



From etienne at isr.ist.utl.pt  Fri Sep  4 20:36:40 2009
From: etienne at isr.ist.utl.pt (etienne at isr.ist.utl.pt)
Date: Fri, 04 Sep 2009 18:36:40 -0700
Subject: bfgsmin works for a version but not for the other
In-Reply-To: <933ae1930909040429k72bae91eka94f6dac1765c7b9@mail.gmail.com>
References: <933ae1930909040429k72bae91eka94f6dac1765c7b9@mail.gmail.com>
Message-ID: <20090904183640.18194ipf08g8uruo@webmail.isr.ist.utl.pt>


   Hi Marcio,

maybe your (new) octave and (older) octave-forge versions don't match?

   Hth,

   Etienne

Quoting M?rcio Diniz <marcio.alves.diniz at gmail.com> (Fri, 4 Sep 2009
08:29:12 -0300):

> Hi there!
> I am using Octave 2.9.9 and the bfgsmin works well. However, the
> same code does not work when I try to run it in Octave 3.0.1 - the
> error message is "warning: bfgsmin: failure, exiting. Try different start
> values?" even though I am using the same starting values.
> Is there any difference between the bfgsmin versions? If not, what
> is the problem?
> Thanks in advance
> Marcio Diniz

-- 
http://www.isr.ist.utl.pt/~etienne

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


From babelproofreader at gmail.com  Sat Sep  5 10:09:36 2009
From: babelproofreader at gmail.com (babelproofreader)
Date: Sat, 5 Sep 2009 08:09:36 -0700 (PDT)
Subject: Moving polynomial fit
Message-ID: <25309400.post@talk.nabble.com>


I would like to write a script/function for use on a time series that will
least squares fit a polynomial of a given degree to a moving window across
the time series in the same way as a moving average computes an average of a
moving window, the output of the script/function being a vector of the
extreme right hand edge/last values of the fitted polynomial over the moving
window. I have tried using the Savitsky-Golay filters from the Signals
package but they are not suitable for my purposes as the end conditions mean
that the values of the fitted polynomial at the right hand edges of the
moving window are different from the values calculated when the data are no
longer at the right hand edge. What in-built Octave functions should I
consider using to achieve the above i.e. polyfit or some other function(s)?
Also, as loops can be quite slow, is there a more efficient way of coding
the above rather than having to resort to a for loop?


-- 
View this message in context: http://www.nabble.com/Moving-polynomial-fit-tp25309400p25309400.html
Sent from the Octave - General mailing list archive at Nabble.com.


From macy at sfo.com  Sat Sep  5 10:38:38 2009
From: macy at sfo.com (macy at sfo.com)
Date: Sat, 5 Sep 2009 08:38:38 -0700 (PDT)
Subject: Moving polynomial fit
In-Reply-To: <25309400.post@talk.nabble.com>
References: <25309400.post@talk.nabble.com>
Message-ID: <1161.66.81.51.141.1252165118.squirrel@cp01.sfo.com>

I had to do something like this for 2 minutes of streaming data once.  The
FOR loop took 20 minutes+ and the array process took 12 seconds!

It went something like this:

Multiply the original vector in such a way as to make a 'shifted' array.

Where each line of the array is your original vector, but shifted slightly.

Then, apply to the array in a single stomp and you should be there

I had to expand the memory octave used, else kept crashing my system when
the data files were up around 20MB

Robert

> I would like to write a script/function for use on a time series that will
> least squares fit a polynomial of a given degree to a moving window across
> the time series in the same way as a moving average computes an average of
> a
> moving window, the output of the script/function being a vector of the
> extreme right hand edge/last values of the fitted polynomial over the
> moving
> window. I have tried using the Savitsky-Golay filters from the Signals
> package but they are not suitable for my purposes as the end conditions
> mean
> that the values of the fitted polynomial at the right hand edges of the
> moving window are different from the values calculated when the data are
> no
> longer at the right hand edge. What in-built Octave functions should I
> consider using to achieve the above i.e. polyfit or some other
> function(s)?
> Also, as loops can be quite slow, is there a more efficient way of coding
> the above rather than having to resort to a for loop?
>
>
> --
> View this message in context:
> http://www.nabble.com/Moving-polynomial-fit-tp25309400p25309400.html
> Sent from the Octave - General mailing list archive at Nabble.com.



From bpabbott at mac.com  Sat Sep  5 12:23:11 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Sat, 05 Sep 2009 13:23:11 -0400
Subject: Moving polynomial fit
In-Reply-To: <25309400.post@talk.nabble.com>
References: <25309400.post@talk.nabble.com>
Message-ID: <A744E98D-CBBF-4A8D-B34B-DBCE89BF95AF@mac.com>

On Sep 5, 2009, at 11:09 AM, babelproofreader wrote:

> I would like to write a script/function for use on a time series  
> that will
> least squares fit a polynomial of a given degree to a moving window  
> across
> the time series in the same way as a moving average computes an  
> average of a
> moving window, the output of the script/function being a vector of the
> extreme right hand edge/last values of the fitted polynomial over  
> the moving
> window. I have tried using the Savitsky-Golay filters from the Signals
> package but they are not suitable for my purposes as the end  
> conditions mean
> that the values of the fitted polynomial at the right hand edges of  
> the
> moving window are different from the values calculated when the data  
> are no
> longer at the right hand edge. What in-built Octave functions should I
> consider using to achieve the above i.e. polyfit or some other  
> function(s)?
> Also, as loops can be quite slow, is there a more efficient way of  
> coding
> the above rather than having to resort to a for loop?

There might be a better way ... Assuming you'd like to center your  
window about the data point of interest ...

---------------- begin: mwpolysmooth.m ---------------
function ym = mwpolysmooth (x, y, order, window)
% USAGE: ym = mwpolysmooth (x, y, order, window)

   xc = num2cell(x);
   idx = @(xc) find (abs (x - xc) < 0.5*window);
   n = cellfun (idx, xc, 'uniformoutput', false);
   mpfit = @(n) polyfit (x(n), y(n), order);
   pm = cellfun (mpfit, n, 'uniformoutput', false);
   nc = num2cell (1:numel(x));
   mpval = @(n) polyval (pm{n}, x(n));
   ym = cell2mat (cellfun (mpval, nc, 'uniformoutput', false));

end

%!demo
%! x = 0:0.1:10;
%! y = rand (size(x)) + cos (x);
%! window = 5;
%! order = 2;
%! ym = mwpolysmooth (x, y, order, window);
%! clf
%! plot (x, y, 's', x, ym)
---------------- end: mwpolysmooth.m ---------------

Ben





From humboldt.fleischer at gmail.com  Sat Sep  5 14:09:02 2009
From: humboldt.fleischer at gmail.com (vHF)
Date: Sat, 5 Sep 2009 12:09:02 -0700 (PDT)
Subject: Supplying Jacobian to lsode
Message-ID: <25311485.post@talk.nabble.com>


Hi all,

I am relatively new to Octave and my problem is most likely rather trivial,
so I apologise in advance.
I am trying to create an .m file that would have a definition of a system of
ODEs and call the lsode solver with the Jacobian supplied. What I have now
is:

***begin script***
u = linspace(0, 72, 72*60);
init = [0; 10];

function res = f(x, u)
  res(1) = 2.0*x(2).*x(2) - 3.0*x(1); 
  res(2) = -2.0*x(2).*x(2) + 6.0*x(1); 
endfunction;

function res = jac(x, u)
  res(1) = -3.0;
res(1,2) = 2.0*x(2);
  res(2,1) = 6.0;
res(2,2) = -2.0*x(2);
endfunction;

x = lsode(["f"; "jac"], init, u);
***end script***

However, this fails with the following output:

***begin console output***
error: `x' undefined near line 5 column 16
error: evaluating binary operator `*' near line 5, column 15
error: evaluating binary operator `.*' near line 5, column 20
error: evaluating binary operator `-' near line 5, column 27
error: evaluating assignment expression near line 5, column 10
error: called from `f'
error: evaluating assignment expression near line 16, column 40
error: called from `__lsode_fcn__U__'
error: lsode: evaluation of user-supplied function failed
error: lsode: inconsistent sizes for state and derivative vectors
error: evaluating assignment expression near line 16, column 3
error: near line 16 of file `/home/marek/Desktop/octave/dimer.m'
***end console output***


What am I doing wrong?

Thanks in advance,

vHF
-- 
View this message in context: http://www.nabble.com/Supplying-Jacobian-to-lsode-tp25311485p25311485.html
Sent from the Octave - General mailing list archive at Nabble.com.


From humboldt.fleischer at gmail.com  Sat Sep  5 14:14:16 2009
From: humboldt.fleischer at gmail.com (vHF)
Date: Sat, 5 Sep 2009 12:14:16 -0700 (PDT)
Subject: Supplying Jacobian to lsode
In-Reply-To: <25311485.post@talk.nabble.com>
References: <25311485.post@talk.nabble.com>
Message-ID: <25311528.post@talk.nabble.com>



vHF wrote:
> 
>   res(1) = -3.0;
> 

This obviously reads "res(1,1) = -3.0" in my script file.

vHF
-- 
View this message in context: http://www.nabble.com/Supplying-Jacobian-to-lsode-tp25311485p25311528.html
Sent from the Octave - General mailing list archive at Nabble.com.


From sraevsky at gmail.com  Sun Sep  6 00:01:41 2009
From: sraevsky at gmail.com (Scott Raevsky)
Date: Sat, 5 Sep 2009 22:01:41 -0700
Subject: Please help me change the prompt
Message-ID: <b6a9929f0909052201g2caccb70sa3f73d25bb7d35e@mail.gmail.com>

Hi,

I just installed octave on Windows.  Whenever I enter something into the
prompt, I get the correct response and then on a newline the following
string: "octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin".  How do I stop
this from printing to the screen every time I hit enter?  It's really
annoying!

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090905/85b75af8/attachment.html 

From rcrummett at gmail.com  Sun Sep  6 01:04:34 2009
From: rcrummett at gmail.com (Ron Crummett)
Date: Sat, 5 Sep 2009 23:04:34 -0700
Subject: Please help me change the prompt
In-Reply-To: <b6a9929f0909052201g2caccb70sa3f73d25bb7d35e@mail.gmail.com>
References: <b6a9929f0909052201g2caccb70sa3f73d25bb7d35e@mail.gmail.com>
Message-ID: <d27c81020909052304p50ae2072t274947e084e57d6b@mail.gmail.com>

create a file .octaverc in your starting directory with the following line:

PS1 ("(your prompt here)");

for example, if you want your prompt to mimic the Matlab prompt, try PS1
(">> ");

The .octaverc file can be used for other customizations as well.

Hope it helps,

-Ron

On Sat, Sep 5, 2009 at 10:01 PM, Scott Raevsky <sraevsky at gmail.com> wrote:

> Hi,
>
> I just installed octave on Windows.  Whenever I enter something into the
> prompt, I get the correct response and then on a newline the following
> string: "octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin".  How do I stop
> this from printing to the screen every time I hit enter?  It's really
> annoying!
>
> Thanks!
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090905/89d56fe4/attachment.html 

From jwe at octave.org  Sun Sep  6 01:22:28 2009
From: jwe at octave.org (John W. Eaton)
Date: Sun, 6 Sep 2009 02:22:28 -0400
Subject: Supplying Jacobian to lsode
In-Reply-To: <25311485.post@talk.nabble.com>
References: <25311485.post@talk.nabble.com>
Message-ID: <19107.21796.715766.482519@segfault.lan>

On  5-Sep-2009, vHF wrote:

| I am relatively new to Octave and my problem is most likely rather trivial,
| so I apologise in advance.
| I am trying to create an .m file that would have a definition of a system of
| ODEs and call the lsode solver with the Jacobian supplied. What I have now
| is:
| 
| ***begin script***
| u = linspace(0, 72, 72*60);
| init = [0; 10];
| 
| function res = f(x, u)
|   res(1) = 2.0*x(2).*x(2) - 3.0*x(1); 
|   res(2) = -2.0*x(2).*x(2) + 6.0*x(1); 
| endfunction;
| 
| function res = jac(x, u)
|   res(1) = -3.0;
| res(1,2) = 2.0*x(2);
|   res(2,1) = 6.0;
| res(2,2) = -2.0*x(2);
| endfunction;
| 
| x = lsode(["f"; "jac"], init, u);
| ***end script***
| 
| However, this fails with the following output:
| 
| ***begin console output***
| error: `x' undefined near line 5 column 16
| error: evaluating binary operator `*' near line 5, column 15
| error: evaluating binary operator `.*' near line 5, column 20
| error: evaluating binary operator `-' near line 5, column 27
| error: evaluating assignment expression near line 5, column 10
| error: called from `f'
| error: evaluating assignment expression near line 16, column 40
| error: called from `__lsode_fcn__U__'
| error: lsode: evaluation of user-supplied function failed
| error: lsode: inconsistent sizes for state and derivative vectors
| error: evaluating assignment expression near line 16, column 3
| error: near line 16 of file `/home/marek/Desktop/octave/dimer.m'
| ***end console output***
| 
| 
| What am I doing wrong?

Try

  x = lsode ({@f, @jac}, init, u);

instead.

I think there is an error in your Jacobian.  I think it should be

  -3   4*x(2)
   6  -4*x(2)

It would probably be slightly more efficient to write

  function res = f (x, u)
    res = zeros (2, 1);
    tmp = 2*x(2)^2;
    res(1) = tmp - 3.0*x(1); 
    res(2) = -tmp + 6.0*x(1); 
  endfunction;

  function res = jac (x, u)
    res = zeros (2, 2);
    tmp = 4.0*x(2);
    res(1,1) = -3.0;
    res(1,2) = tmp;
    res(2,1) = 6.0;
    res(2,2) = -tmp;
  endfunction;

or perhaps

 function res = f (x, u)
   tmp = 2*x(2)^2;
   res = [tmp-3*x(1); -tmp+6*x(1)];
 endfunction
 
 function res = jac (x, u)
   tmp = 4*x(2);
   res = [-3, tmp; 6, -tmp];
 endfunction

because your functions are repeatedly resizing the res vectors.

BTW, this system appears to be unstable, and all the interesting
behavior seems to happen before u = 1.

jwe

From tmacchant at yahoo.co.jp  Sun Sep  6 03:59:53 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Sun, 6 Sep 2009 17:59:53 +0900 (JST)
Subject: Fwd: Re: Please help me change the prompt
Message-ID: <20090906085953.16623.qmail@web3811.mail.bbt.yahoo.co.jp>



--- Tatsuro MATSUOKA <tmacchant at yahoo.co.jp> wrote:

> Date:Sun, 6 Sep 2009 15:10:10 +0900 (JST)
> From:Tatsuro MATSUOKA <tmacchant at yahoo.co.jp>
> Subject:Re: Please help me change the prompt
> To:Scott Raevsky <sraevsky at gmail.com>
> 
> Hello
> Please comment out line 74 of .....\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\startup\octaverc
> 
> 72: ## set default prompt
> 73:
> 74: PS1("\\s:\\#:\\w\n> ");
> |
> V
> 74 # PS1("\\s:\\#:\\w\n> ");
> 
> or change like 
> 74: PS1(">> ");
> 
> Regards
> 
> Tatsuro 
> --- Scott Raevsky wrote:
> 
> > Hi,
> > 
> > I just installed octave on Windows.  Whenever I enter something into the
> > prompt, I get the correct response and then on a newline the following
> > string: "octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin".  How do I stop
> > this from printing to the screen every time I hit enter?  It's really
> > annoying!
> > 
> > Thanks!
> > > _______________________________________________
> > Help-octave mailing list
> > Help-octave at octave.org
> > https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> > 
> 
> 
> --------------------------------------
> Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
> http://pr.mail.yahoo.co.jp/ec10years/
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From humboldt.fleischer at gmail.com  Sun Sep  6 06:13:19 2009
From: humboldt.fleischer at gmail.com (vHF)
Date: Sun, 6 Sep 2009 04:13:19 -0700 (PDT)
Subject: Supplying Jacobian to lsode
In-Reply-To: <19107.21796.715766.482519@segfault.lan>
References: <25311485.post@talk.nabble.com>
	<19107.21796.715766.482519@segfault.lan>
Message-ID: <25316691.post@talk.nabble.com>


Dear John,


John W. Eaton-3 wrote:
> 
> Try
> 
>   x = lsode ({@f, @jac}, init, u);
> 
> instead.
> 

Brilliant -- this works.


John W. Eaton wrote:
> 
> I think there is an error in your Jacobian.  
> ...
> BTW, this system appears to be unstable, and all the interesting
> behavior seems to happen before u = 1.
> 

I should have mentioned that I submitted the smallest example 
I could think of, not the real thing. Now I am going back to 60+ 
equations, so your performance tips are also going to be very useful!

Thanks,

Marek (vHF)

-- 
View this message in context: http://www.nabble.com/Supplying-Jacobian-to-lsode-tp25311485p25316691.html
Sent from the Octave - General mailing list archive at Nabble.com.


From febty82 at gmail.com  Sun Sep  6 23:25:31 2009
From: febty82 at gmail.com (febty febriani)
Date: Mon, 7 Sep 2009 13:25:31 +0900
Subject: how to change scale display in imagesc
Message-ID: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>

Dear all,

I plotted the attached figure with using imagesc. My question is how to
change the scale display so that the y- scale from 0 to 0.5, not from 0.5 to
0 like the attached figure. Any thoughts are appreciated. Thanks very much..


Your sincerely,
Febty
-- 

******
febty febriani
Indonesian Institute of Sciences
Research Center for Physics
Kompleks PUSPIPTEK
Serpong, Indonesia, 15314
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/0e3c8385/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: output20090101.png
Type: image/png
Size: 25746 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/0e3c8385/attachment-0001.png 

From febty82 at gmail.com  Mon Sep  7 00:17:32 2009
From: febty82 at gmail.com (febty febriani)
Date: Mon, 7 Sep 2009 14:17:32 +0900
Subject: Fwd: how to change scale display in imagesc
In-Reply-To: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>
References: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>
Message-ID: <1964ddb30909062217g2e3744ebwa24693bcc943f740@mail.gmail.com>

Dear all,

I plotted the attached figure with using imagesc. My question is how to
change the scale display so that the y- scale from 0 to 0.5, not from 0.5 to
0 like the attached figure. Any thoughts are appreciated. Thanks very much..


Your sincerely,
Febty
-- 

******
febty febriani
Indonesian Institute of Sciences
Research Center for Physics
Kompleks PUSPIPTEK
Serpong, Indonesia, 15314
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/73b28ae8/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: output20090101.png
Type: image/png
Size: 25746 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/73b28ae8/attachment-0001.png 

From michael.creel at uab.es  Mon Sep  7 04:40:10 2009
From: michael.creel at uab.es (Michael Creel)
Date: Mon, 7 Sep 2009 02:40:10 -0700 (PDT)
Subject: bfgsmin works for a version but not for the other
In-Reply-To: <933ae1930909040429k72bae91eka94f6dac1765c7b9@mail.gmail.com>
References: <933ae1930909040429k72bae91eka94f6dac1765c7b9@mail.gmail.com>
Message-ID: <25327709.post@talk.nabble.com>




M?rcio Diniz wrote:
> 
> Hi there!
> I am using Octave 2.9.9 and the bfgsmin works well. However, the same code
> does not work when I try to run it in Octave 3.0.1 - the error message is
> "warning: bfgsmin: failure, exiting. Try different start values?" even
> though I am using the same starting values.
> Is there any difference between the bfgsmin versions? If not, what is the
> problem?
> Thanks in advance
> Marcio Diniz
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

Hi,
The version of __bfgsmin.cc in Octave Forge had some problem, a fixed
version was checked in this morning. You can get it at 
http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/main/optim/src/__bfgsmin.cc?revision=6229&view=markup
Just copy it over the version you have, and execute "mkoctfile __bfgsmin.cc"
M.

-- 
View this message in context: http://www.nabble.com/bfgsmin-works-for-a-version-but-not-for-the-other-tp25294160p25327709.html
Sent from the Octave - General mailing list archive at Nabble.com.



From ct529 at york.ac.uk  Mon Sep  7 07:38:28 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 7 Sep 2009 13:38:28 +0100
Subject: Matrix regression of distnce matrices + non negative least square
Message-ID: <200909071338.28862.ct529@york.ac.uk>

Dear friends,

I would like to solve the following regression problem:

y=c1 x1 + c2 x2 + .... + cn xn

where the y, xi are all matrices and the ci are constants that need to be 
determined. The y, xi are distance matrices (symmetric). Obviously ci should 
be forced to positive or null (i.e. non negative).

Any suggestion? 

I will be more than happy to share the results of my quest with the list or 
developers.
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Mon Sep  7 08:39:04 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 7 Sep 2009 15:39:04 +0200
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <200909071338.28862.ct529@york.ac.uk>
References: <200909071338.28862.ct529@york.ac.uk>
Message-ID: <69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>

On Mon, Sep 7, 2009 at 2:38 PM, Corrado<ct529 at york.ac.uk> wrote:
> Dear friends,
>
> I would like to solve the following regression problem:
>
> y=c1 x1 + c2 x2 + .... + cn xn
>
> where the y, xi are all matrices and the ci are constants that need to be
> determined. The y, xi are distance matrices (symmetric). Obviously ci should
> be forced to positive or null (i.e. non negative).
>
> Any suggestion?
>

First convert the linear system into canonical form (matrix of
coefficients and vector of rhs), then use lsqnonneg.
Since the matrices are symmetric, extracting just lower/upper
triangular parts will suffice. The best way how to do the first will
depend on the dimensionality of your problem and how your data is
stored.

regards
-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Mon Sep  7 09:21:39 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 7 Sep 2009 15:21:39 +0100
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>
References: <200909071338.28862.ct529@york.ac.uk>
	<69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>
Message-ID: <200909071521.39612.ct529@york.ac.uk>

Dear Jaroslav,

thanks for your help.

I actually do not understand what you are suggesting:

1) The data would be stored in matrices, which I can easily dump into csv files
2) The distance matrices are up to 5000 x 5000 (but they symmetric, so only 
n(n-1)/2 are important), but the ones I am using now are just 1500 x 1500.
3) I thought lsqnonneg only worked on linear system in the form y=ax
4) The terms are matrices, how do you write a canonical form???? the x1 .... 
xn are matrices (e.g. 1500 x 1500) .... the ci are coefficients in R ....

Regards

On Monday 07 September 2009 14:39:04 you wrote:
> On Mon, Sep 7, 2009 at 2:38 PM, Corrado<ct529 at york.ac.uk> wrote:
> > Dear friends,
> >
> > I would like to solve the following regression problem:
> >
> > y=c1 x1 + c2 x2 + .... + cn xn
> >
> > where the y, xi are all matrices and the ci are constants that need to be
> > determined. The y, xi are distance matrices (symmetric). Obviously ci
> > should be forced to positive or null (i.e. non negative).
> >
> > Any suggestion?
>
> First convert the linear system into canonical form (matrix of
> coefficients and vector of rhs), then use lsqnonneg.
> Since the matrices are symmetric, extracting just lower/upper
> triangular parts will suffice. The best way how to do the first will
> depend on the dimensionality of your problem and how your data is
> stored.
>
> regards



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From babelproofreader at gmail.com  Mon Sep  7 09:27:55 2009
From: babelproofreader at gmail.com (babelproofreader)
Date: Mon, 7 Sep 2009 07:27:55 -0700 (PDT)
Subject: Moving polynomial fit
In-Reply-To: <A744E98D-CBBF-4A8D-B34B-DBCE89BF95AF@mac.com>
References: <25309400.post@talk.nabble.com>
	<A744E98D-CBBF-4A8D-B34B-DBCE89BF95AF@mac.com>
Message-ID: <25331404.post@talk.nabble.com>



On Sep 5, 2009, at 11:09 AM, babelproofreader wrote:

> I would like to write a script/function for use on a time series  
> that will
> least squares fit a polynomial of a given degree to a moving window  
> across
> the time series in the same way as a moving average computes an  
> average of a
> moving window.....

There might be a better way ... Assuming you'd like to center your  
window about the data point of interest ...

---------------- begin: mwpolysmooth.m ---------------
function ym = mwpolysmooth (x, y, order, window)
% USAGE: ym = mwpolysmooth (x, y, order, window)

   xc = num2cell(x);
   idx = @(xc) find (abs (x - xc) < 0.5*window);
   n = cellfun (idx, xc, 'uniformoutput', false);
   mpfit = @(n) polyfit (x(n), y(n), order);
   pm = cellfun (mpfit, n, 'uniformoutput', false);
   nc = num2cell (1:numel(x));
   mpval = @(n) polyval (pm{n}, x(n));
   ym = cell2mat (cellfun (mpval, nc, 'uniformoutput', false));

end
---------------- end: mwpolysmooth.m ---------------

Ben,

Thanks for your function - I've learned some useful coding from it - but I'm
not sure if it really address my problem. I wish to fit real time streaming
data (market prices), but as I understand it your code "peaks into the
future" to smooth over the window i.e. the fitted value for the most recent
streaming data "n" is different from the value that will be calculated when
this data value is (n - some_value) in the past, hence invalidating any
historical testing over past prices. The simple for loop below is an example
of what I want to do, but it is slow.

x=(1:length(price))';
polysmooth=price;
for i=5:length(price)
y=price(i-4:i,1);
a=polyfit(x,y,2);
polysmooth(i,1)=polyval(a,5);
endfor






-- 
View this message in context: http://www.nabble.com/Moving-polynomial-fit-tp25309400p25331404.html
Sent from the Octave - General mailing list archive at Nabble.com.


From ct529 at york.ac.uk  Mon Sep  7 09:37:08 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 7 Sep 2009 15:37:08 +0100
Subject: Matrix regression of distance matrices + non negative least square
In-Reply-To: <69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>
References: <200909071338.28862.ct529@york.ac.uk>
	<69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>
Message-ID: <200909071537.08610.ct529@york.ac.uk>

Dear Jaroslav,

by the way, 

lsqnonneg

is not among the packages I can install on my octave. I know there is a 
lsqnonneg package in Matlab.

Regard

On Monday 07 September 2009 14:39:04 Jaroslav Hajek wrote:
> On Mon, Sep 7, 2009 at 2:38 PM, Corrado<ct529 at york.ac.uk> wrote:
> > Dear friends,
> >
> > I would like to solve the following regression problem:
> >
> > y=c1 x1 + c2 x2 + .... + cn xn
> >
> > where the y, xi are all matrices and the ci are constants that need to be
> > determined. The y, xi are distance matrices (symmetric). Obviously ci
> > should be forced to positive or null (i.e. non negative).
> >
> > Any suggestion?
>
> First convert the linear system into canonical form (matrix of
> coefficients and vector of rhs), then use lsqnonneg.
> Since the matrices are symmetric, extracting just lower/upper
> triangular parts will suffice. The best way how to do the first will
> depend on the dimensionality of your problem and how your data is
> stored.
>
> regards



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From macy at sfo.com  Mon Sep  7 10:00:40 2009
From: macy at sfo.com (macy at sfo.com)
Date: Mon, 7 Sep 2009 08:00:40 -0700 (PDT)
Subject: Moving polynomial fit
In-Reply-To: <25331404.post@talk.nabble.com>
References: <25309400.post@talk.nabble.com>
	<A744E98D-CBBF-4A8D-B34B-DBCE89BF95AF@mac.com>
	<25331404.post@talk.nabble.com>
Message-ID: <1083.66.81.37.191.1252335640.squirrel@cp01.sfo.com>

NOW I REMEMBER!

I had to rewrite the function polyval into a custom type function so that
I could use c variable as an array, not just a vector.

Then, just write the whole for-loop as an array sequence.  As I said I
went from 20 minutes down to 12 seconds for mine.

Robert

> On Sep 5, 2009, at 11:09 AM, babelproofreader wrote:
>
>> I would like to write a script/function for use on a time series
>> that will
>> least squares fit a polynomial of a given degree to a moving window
>> across
>> the time series in the same way as a moving average computes an
>> average of a
>> moving window.....
>
> There might be a better way ... Assuming you'd like to center your
> window about the data point of interest ...
>
> ---------------- begin: mwpolysmooth.m ---------------
> function ym = mwpolysmooth (x, y, order, window)
> % USAGE: ym = mwpolysmooth (x, y, order, window)
>
>    xc = num2cell(x);
>    idx = @(xc) find (abs (x - xc) < 0.5*window);
>    n = cellfun (idx, xc, 'uniformoutput', false);
>    mpfit = @(n) polyfit (x(n), y(n), order);
>    pm = cellfun (mpfit, n, 'uniformoutput', false);
>    nc = num2cell (1:numel(x));
>    mpval = @(n) polyval (pm{n}, x(n));
>    ym = cell2mat (cellfun (mpval, nc, 'uniformoutput', false));
>
> end
> ---------------- end: mwpolysmooth.m ---------------
>
> Ben,
>
> Thanks for your function - I've learned some useful coding from it - but
> I'm
> not sure if it really address my problem. I wish to fit real time
> streaming
> data (market prices), but as I understand it your code "peaks into the
> future" to smooth over the window i.e. the fitted value for the most
> recent
> streaming data "n" is different from the value that will be calculated
> when
> this data value is (n - some_value) in the past, hence invalidating any
> historical testing over past prices. The simple for loop below is an
> example
> of what I want to do, but it is slow.
>
> x=(1:length(price))';
> polysmooth=price;
> for i=5:length(price)
> y=price(i-4:i,1);
> a=polyfit(x,y,2);
> polysmooth(i,1)=polyval(a,5);
> endfor
>
>



From highegg at gmail.com  Mon Sep  7 10:04:35 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 7 Sep 2009 17:04:35 +0200
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <200909071521.39612.ct529@york.ac.uk>
References: <200909071338.28862.ct529@york.ac.uk>
	<69d8d540909070639w19be493eke4b2d615b5400f2a@mail.gmail.com>
	<200909071521.39612.ct529@york.ac.uk>
Message-ID: <69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>

On Mon, Sep 7, 2009 at 4:21 PM, Corrado<ct529 at york.ac.uk> wrote:
> Dear Jaroslav,
>
> thanks for your help.
>
> I actually do not understand what you are suggesting:
>
> 1) The data would be stored in matrices, which I can easily dump into csv files

OK. But mind you, instead of working with a number of equal-sized
matrices it is often more efficient to work with a 3D array. Usually
it depends on where you get the matrices from.

> 2) The distance matrices are up to 5000 x 5000 (but they symmetric, so only
> n(n-1)/2 are important), but the ones I am using now are just 1500 x 1500.

In that case, it probably doesn't matter much which representation you choose.

> 3) I thought lsqnonneg only worked on linear system in the form y=ax

Yes. That's why you need to bring the system into this form.

> 4) The terms are matrices, how do you write a canonical form???? the x1 ....
> xn are matrices (e.g. 1500 x 1500) .... the ci are coefficients in R ....
>

A matrix equation constitutes a system of equations; one for each
element. If x1...xN are MxM, you get a
system of M*(M+1)/2 equations in N variables.

Suppose you have the matrices X in a cell array, so that X{1} is the
first matrix etc., and the RHS matrix is Y.
Then you want to do something like
A = [cellfun (@vech, X, "UniformOutput", false){:}];
b = vech (Y);
c = lsqnonneg (A, b);
To get the coefficients in "c". Note that lsqnonneg is part of Octave 3.2.0+.
There is a small problem with the above simplistic code: it does not
discard the diagonal equations, because vech includes the diagonal.
You can modify vech in order to discard even the diagonal equations.

hth

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Mon Sep  7 10:13:04 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 7 Sep 2009 16:13:04 +0100
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
References: <200909071338.28862.ct529@york.ac.uk>
	<200909071521.39612.ct529@york.ac.uk>
	<69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
Message-ID: <200909071613.04573.ct529@york.ac.uk>

Thanks a lot Jaroslav, I am trying to understand how it works .... :) .... it 
is my first program in octave.

On Monday 07 September 2009 16:04:35 Jaroslav Hajek wrote:
> On Mon, Sep 7, 2009 at 4:21 PM, Corrado<ct529 at york.ac.uk> wrote:
> > Dear Jaroslav,
> >
> > thanks for your help.
> >
> > I actually do not understand what you are suggesting:
> >
> > 1) The data would be stored in matrices, which I can easily dump into csv
> > files
>
> OK. But mind you, instead of working with a number of equal-sized
> matrices it is often more efficient to work with a 3D array. Usually
> it depends on where you get the matrices from.
>
> > 2) The distance matrices are up to 5000 x 5000 (but they symmetric, so
> > only n(n-1)/2 are important), but the ones I am using now are just 1500 x
> > 1500.
>
> In that case, it probably doesn't matter much which representation you
> choose.
>
> > 3) I thought lsqnonneg only worked on linear system in the form y=ax
>
> Yes. That's why you need to bring the system into this form.
>
> > 4) The terms are matrices, how do you write a canonical form???? the x1
> > .... xn are matrices (e.g. 1500 x 1500) .... the ci are coefficients in R
> > ....
>
> A matrix equation constitutes a system of equations; one for each
> element. If x1...xN are MxM, you get a
> system of M*(M+1)/2 equations in N variables.
>
> Suppose you have the matrices X in a cell array, so that X{1} is the
> first matrix etc., and the RHS matrix is Y.
> Then you want to do something like
> A = [cellfun (@vech, X, "UniformOutput", false){:}];
> b = vech (Y);
> c = lsqnonneg (A, b);
> To get the coefficients in "c". Note that lsqnonneg is part of Octave
> 3.2.0+. There is a small problem with the above simplistic code: it does
> not discard the diagonal equations, because vech includes the diagonal. You
> can modify vech in order to discard even the diagonal equations.
>
> hth



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From bpabbott at mac.com  Mon Sep  7 10:18:06 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Mon, 07 Sep 2009 11:18:06 -0400
Subject: Moving polynomial fit
In-Reply-To: <25331404.post@talk.nabble.com>
References: <25309400.post@talk.nabble.com>
	<A744E98D-CBBF-4A8D-B34B-DBCE89BF95AF@mac.com>
	<25331404.post@talk.nabble.com>
Message-ID: <F3EFD226-261B-4656-A0FD-046A828E0970@mac.com>


On Sep 7, 2009, at 10:27 AM, babelproofreader wrote:

>
>
> On Sep 5, 2009, at 11:09 AM, babelproofreader wrote:
>
>> I would like to write a script/function for use on a time series
>> that will
>> least squares fit a polynomial of a given degree to a moving window
>> across
>> the time series in the same way as a moving average computes an
>> average of a
>> moving window.....
>
> There might be a better way ... Assuming you'd like to center your
> window about the data point of interest ...
>
> ---------------- begin: mwpolysmooth.m ---------------
> function ym = mwpolysmooth (x, y, order, window)
> % USAGE: ym = mwpolysmooth (x, y, order, window)
>
>   xc = num2cell(x);
>   idx = @(xc) find (abs (x - xc) < 0.5*window);
>   n = cellfun (idx, xc, 'uniformoutput', false);
>   mpfit = @(n) polyfit (x(n), y(n), order);
>   pm = cellfun (mpfit, n, 'uniformoutput', false);
>   nc = num2cell (1:numel(x));
>   mpval = @(n) polyval (pm{n}, x(n));
>   ym = cell2mat (cellfun (mpval, nc, 'uniformoutput', false));
>
> end
> ---------------- end: mwpolysmooth.m ---------------
>
> Ben,
>
> Thanks for your function - I've learned some useful coding from it -  
> but I'm
> not sure if it really address my problem. I wish to fit real time  
> streaming
> data (market prices), but as I understand it your code "peaks into the
> future" to smooth over the window i.e. the fitted value for the most  
> recent
> streaming data "n" is different from the value that will be  
> calculated when
> this data value is (n - some_value) in the past, hence invalidating  
> any
> historical testing over past prices.


You can modify the window to represent only past values by ...

   idx = @(xc) find ((xc - x) < window & (xc - x) > 0);

Ben




From xdong at cse.unl.edu  Mon Sep  7 10:33:32 2009
From: xdong at cse.unl.edu (Xin Dong)
Date: Mon, 7 Sep 2009 10:33:32 -0500
Subject: Greek letters in plot legend
Message-ID: <7a394d4e0909070833w107cfc2dj100d67eb99232e7f@mail.gmail.com>

Hi all,

I need Greek letters in plot legend. I use the command:
legend("\eta=5", "\eta=10", "\eta=15", "\eta=20", "location", "southwest");

However, octave cannot parse "\eta". The interesting thing is if I use
"\alpha" or "\theta", it can be successfully displayed. I tried it on Octave
3.0.1 on Unbuntu 9.04 and Octave 3.2.2 on Mac OS X, both have the same
problem.

Can anyone give some advice?

Thanks,
Xin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/3fe7558e/attachment.html 

From bpabbott at mac.com  Mon Sep  7 10:55:51 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Mon, 07 Sep 2009 11:55:51 -0400
Subject: Greek letters in plot legend
In-Reply-To: <7a394d4e0909070833w107cfc2dj100d67eb99232e7f@mail.gmail.com>
References: <7a394d4e0909070833w107cfc2dj100d67eb99232e7f@mail.gmail.com>
Message-ID: <5895F928-0B9F-40ED-870F-A6368737A679@mac.com>


On Sep 7, 2009, at 11:33 AM, Xin Dong wrote:

> Hi all,
>
> I need Greek letters in plot legend. I use the command:
> legend("\eta=5", "\eta=10", "\eta=15", "\eta=20", "location",  
> "southwest");
>
> However, octave cannot parse "\eta". The interesting thing is if I  
> use "\alpha" or "\theta", it can be successfully displayed. I tried  
> it on Octave 3.0.1 on Unbuntu 9.04 and Octave 3.2.2 on Mac OS X,  
> both have the same problem.
>
> Can anyone give some advice?
>
> Thanks,
> Xin


See this page ..

	http://www.gnu.org/software/octave/doc/interpreter/Strings.html

The backslash is being interpreted as an escape sequence. What you  
need to correct it is ...

	legend("\\eta=5", "\\eta=10", "\\eta=15", "\\eta=20", "location",  
"southwest");

... or replace the double-quotes with single-quotes ...

	legend('\eta=5', '\eta=10', '\eta=15', '\eta=20', "location",  
"southwest");

Ben


From ilnseher at eit.uni-kl.de  Mon Sep  7 11:04:53 2009
From: ilnseher at eit.uni-kl.de (Thomas Ilnseher)
Date: Mon, 07 Sep 2009 18:04:53 +0200
Subject: [Fwd: Re: Greek letters in plot legend]
Message-ID: <1252339493.1971.57.camel@pcmik05.zmk.uni-kl.de>

I forget to send to the octave ML.
-- 
Thomas Ilnseher <ilnseher at eit.uni-kl.de>
-------------- next part --------------
An embedded message was scrubbed...
From: Thomas Ilnseher <ilnseher at eit.uni-kl.de>
Subject: Re: Greek letters in plot legend
Date: Mon, 07 Sep 2009 18:03:53 +0200
Size: 27525
Url: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/6857506a/attachment-0001.mht 

From carlo.defalco at gmail.com  Mon Sep  7 12:25:33 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Mon, 7 Sep 2009 19:25:33 +0200
Subject: [Fwd: Re: Greek letters in plot legend]
In-Reply-To: <1252339493.1971.57.camel@pcmik05.zmk.uni-kl.de>
References: <1252339493.1971.57.camel@pcmik05.zmk.uni-kl.de>
Message-ID: <B9E52EC5-3C77-4D80-9A89-3CBFB1620F32@gmail.com>


On 7 Sep 2009, at 18:04, Thomas Ilnseher wrote:

> depends on what you need to do.
>
> If you want to see it on the monitor, i can't help you.
>
> If you need it in files, I have an advise for you:
>
> octave:1> x=0:0.01:2*pi;
> octave:2> y=sin(x);
> ^^ I plot something
> octave:3> plot(x,y,'.-b;\$sin(\eta)$;');
>                          ^^ I use latex stuff for the greek letters -
> gnuplot complains

the following works fine for me with octave 3.2.2, gnuplot 4.2.5 and  
X11 display on a mac:

 >> x=0:0.01:2*pi;y=sin(x);
 >> plot(x,y);
 >> legend('\sin(\eta)')

HTH,
c.

From Rolf.Schirmacher at MuellerBBM.de  Mon Sep  7 14:31:11 2009
From: Rolf.Schirmacher at MuellerBBM.de (Schirmacher, Rolf)
Date: Mon, 7 Sep 2009 21:31:11 +0200 
Subject: bandpass filter for 2D Data
Message-ID: <2417729F3416D511A27E0090273A963B02910FD1@smex01.mbbm.de>



> -----Original Message-----
> From: John Swensen [mailto:jpswensen at comcast.net]
> Sent: Friday, August 28, 2009 5:33 PM
> To: Sergei Steshenko
> Cc: help-octave at octave.org
> Subject: Re: bandpass filter for 2D Data
> 
> 
> 
> On Aug 28, 2009, at 11:30 AM, Sergei Steshenko wrote:
> 
> >
> >
> > --- On Thu, 8/27/09, John Swensen <jpswensen at comcast.net> wrote:
> >
> >> From: John Swensen <jpswensen at comcast.net>
> >> Subject: Re: bandpass filter for 2D Data
> >> To: "Dr. Johannes Zellner" <johannes at zellner.org>
> >> Cc: help-octave at octave.org
> >> Date: Thursday, August 27, 2009, 12:58 PM
> >>
> >> On Aug 27, 2009, at 3:15 AM, Dr. Johannes Zellner wrote:
> >>
> >>> Hi
> >>>
> >>> does anyone have an implementation of a 2D bandpass
> >> filter?
> >>>
> >>> Suppose I've 2d data in a matrix of 1000 x 1000 pixels
> >> which
> >>> represents a spatial extension of 1 m (1 mm per
> >> pixel).
> >>> Now I'd like to filter away all frequencies below a
> >> wavelength of 50
> >>> mm and above a wavelength of 150 mm,
> >>> so the resulting data should only have frequencies
> >> corresponding to
> >>> wavelengths between 50 and 150 mm.
> >>>
> >>> Any octave examples for this?
> >>>
> >>> --   
> >>> Johannes
> >>> _______________________________________________
> >>> Help-octave mailing list
> >>> Help-octave at octave.org
> >>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> >>
> >> I don't know if there is an example of how to do this, but
> >> I think I
> >> remember doing the following pseudocode in my computer
> >> visions class:
> >>
> >> 1) perform the 2D FFT
> >> 2) figure out, based on your spatial sampling frequency,
> >> how much one
> >> "pixel" in the magnitude portion of the 2D FFT is
> >> 3) set everything but an annulus whose larger radius
> >> equates to 150mm
> >> wavelength and inner radius equates to 50mm to zero.
> >> 4) Perform the inverse 2D FFT
> >>
> >> Someone please correct me if this doesn't work, but I think
> >> this ends
> >> up being a fairly quick way to do what you ask and easy to
> >> implement.
> >>
> >> John Swensen
> >> _______________________________________________
> >
> >
> > Your approach is essentially correct, however, with any inverse FFT
> > brickwall filter (the one you propose) causes a lot of 
> ringing in the
> > reconstituted signal.
> >
> > Regards,
> >  Sergei.
> >
> >
> >
> 
> Yeah, I thought about that afterward, but hadn't got around to  
> correcting myself.  To solve that problem you use a difference of 2D  
> Guassians to generate the annulus.  I don't quite remember 
> now how to  
> set the variance of the Gaussians to get the desired cutoff  
> frequencies, though.
> 
> Thanks for pointing out my error,
> John
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 

I guess the same way as in 1d filter design... and unfortunately the optimum
solution depends on your performance criterion. So, perhaps do a 1d - design
and "circle" the frequency response (complex!) around zero (probably) in the
complex spectral plane to get the spectral filter matrix. Gaussian might be
preferred due to its optimum time-frequency properties (i.e. equally
distributed time-frequency uncertainty / fourier-transform is gauss again),
but that really depends on the application. Getting a frequency domain brick
of course gives about the worst time/spatial domain behaviour.

Rolf Schirmacher

From christophe.tournery at illusonic.com  Mon Sep  7 16:24:29 2009
From: christophe.tournery at illusonic.com (Christophe Tournery)
Date: Mon, 7 Sep 2009 23:24:29 +0200
Subject: how to change scale display in imagesc
In-Reply-To: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>
References: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>
Message-ID: <94A29D38-8E28-4F2F-B48D-31CC7BE3A71B@illusonic.com>

You may want to try:
axis xy
-c

On Sep 7, 2009, at 6:25 AM, febty febriani wrote:

> Dear all,
>
> I plotted the attached figure with using imagesc. My question is how  
> to change the scale display so that the y- scale from 0 to 0.5, not  
> from 0.5 to 0 like the attached figure. Any thoughts are  
> appreciated. Thanks very much..
>
>
> Your sincerely,
> Febty
> -- 
>
> ******
> febty febriani
> Indonesian Institute of Sciences
> Research Center for Physics
> Kompleks PUSPIPTEK
> Serpong, Indonesia, 15314
> <output20090101.png>_______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave


From kirzhanov at gmail.com  Mon Sep  7 10:39:50 2009
From: kirzhanov at gmail.com (iason)
Date: Mon, 7 Sep 2009 08:39:50 -0700 (PDT)
Subject: lsode
In-Reply-To: <25A0E03F32D04501A0BCD48984753C78@cb289178a>
References: <25A0E03F32D04501A0BCD48984753C78@cb289178a>
Message-ID: <25332412.post@talk.nabble.com>



dmelliott wrote:
> 
> 
> Dear Help,
> 
>   Is it possible to get the actual solution points that lsode uses rather
> than just the regularly spaced ones generated by linspace for the input?
> 
>                                                                                            
> Thanks,
> 
>                                                                                                   
> dmelliott
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

dmelliott,

have you achieved any success in retrieving the solution points (the
independent variable nodes) from the LSODE function of OCTAVE? don't you
know, is it possible to do this using the original LSODE (or similar)
function from the ODEPACK library (where it comes from)?

-- 
View this message in context: http://www.nabble.com/lsode-tp19545666p25332412.html
Sent from the Octave - General mailing list archive at Nabble.com.


From gbohanno at netscape.net  Mon Sep  7 18:52:58 2009
From: gbohanno at netscape.net (George Bohannon)
Date: Mon, 07 Sep 2009 19:52:58 -0400
Subject: Trouble building 3.2.2 (pthread)
Message-ID: <8CBFE2D75F48880-28C0-1D9C2@webmail-m034.sysops.aol.com>

I'm trying to build 3.2.2 on a Red Hat EL4 system. I installed gcc 4.3.4 specifically for this purpose. I ran configure successfully as follows:

./configure F77=gfortran

I then ran 'make' and got the following error after building for a while. Does anyone know how to satisfy this undefined reference to 'pthread' items?

gcc -c -I/usr/include/pcre -I. -I.. -I../liboctave -I../src -I../libcruft/misc? -DHAVE_CONFIG_H? -Wall -W -Wshadow -W
format -g -O2 main.c -o main.o
g++ -I/usr/include/pcre -I. -I.. -I../liboctave -I../src -I../libcruft/misc? -DHAVE_CONFIG_H? -Wall -W -Wshadow -Wold
-style-cast -Wformat -g -O2 -rdynamic \
-L..? -fPIC? -o octave \
main.o? \
-L../liboctave -L../libcruft -L../src -Wl,-rpath -Wl,/usr/local/lib/octave-3.2.2 \
-loctinterp -loctave? -lcruft?? \
???? \
??? \
??? \
-L/usr/X11R6/lib64 -lX11? -lreadline? -lncurses -ldl -lblas -lz -lm? -L./../lib64 -L/usr/local/lib/gcc/x86_64-unknown
-linux-gnu/4.3.4 -L/usr/local/lib/gcc/x86_64-unknown-linux-gnu/4.3.4/../../../../lib64 -L/lib/../lib64 -L/usr/lib/../
lib64 -L. -L/usr/local/lib/gcc/x86_64-unknown-linux-gnu/4.3.4/../../.. -lz -lgfortranbegin -lgfortran -lm
../liboctave/liboctave.so: undefined reference to `pthread_mutexattr_destroy'
../liboctave/liboctave.so: undefined reference to `pthread_mutexattr_settype'
../liboctave/liboctave.so: undefined reference to `pthread_mutexattr_init'
collect2: ld returned 1 exit status
make[2]: *** [octave] Error 1
make[2]: Leaving directory `/usr/local/octave-3.2.2/src'
make[1]: *** [src] Error 2
make[1]: Leaving directory `/usr/local/octave-3.2.2'
make: *** [all] Error 2

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090907/cf7f0a9c/attachment.html 

From tmacchant at yahoo.co.jp  Tue Sep  8 02:39:42 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Tue, 8 Sep 2009 16:39:42 +0900 (JST)
Subject: Can I change maker of stem?
Message-ID: <20090908073942.78562.qmail@web3804.mail.bbt.yahoo.co.jp>

Hello

This inquiry is done in the octave thread in Japan.

In the manual

The default color is "r" (red). The default line style is "-" and the default marker is
"o". The line style can be altered by the linespec argument in the same manner as
the plot command. 

However, there is no description of alternation of the marker. 
Can I change maker of stem?  For example, "x".

It will be grateful for me if someone will give me information.

Regards

Tatsuro  

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From schubert.seb at googlemail.com  Tue Sep  8 03:37:16 2009
From: schubert.seb at googlemail.com (Sebastian Schubert)
Date: Tue, 8 Sep 2009 10:37:16 +0200
Subject: Logarithmic hist and other colour
Message-ID: <200909081037.20947.schubert.seb@googlemail.com>

Hi,

is it possible to plot a histogram logarithmically within octave (3.2.2) 
with the standard tools? Can I change the colour (no filling of the 
bars)? Pointers to the manual are fine. :)

Cheers
Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090908/6606b91b/attachment.bin 

From benjamin.barras at epfl.ch  Tue Sep  8 09:40:05 2009
From: benjamin.barras at epfl.ch (alynghy)
Date: Tue, 8 Sep 2009 07:40:05 -0700 (PDT)
Subject: gnuplot not working in RHEL4 32 bit
In-Reply-To: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
References: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
Message-ID: <25347537.post@talk.nabble.com>


The problem is solved on RHEL 5.3 with the same errors :

1) Remove gnuplot from redhat :

yum erase gnuplot

2) Remove libraries blas/lapack from redhat :

yum erase lapack
yum erase blas

3) Remove oldies :

yum erase compat-*
yum erase compat-lib*

4) Compile the last gnuplot and lapack :

cd gnuplot-4.2.5
./configure
make
make install
cd ..

AND

cd lapack-3.2.1
cp make.inc.example make.inc
make blaslib
make all
cp blas_LINUX.a /usr/local/lib/libblas.a
cp lapack_LINUX.a /usr/local/lib/liblapack.a

5) Compile octave :

cd octave-3.2.2
./configure --with-blas=/usr/local/lib  --with-lapack=/usr/local/lib
--enable-static LIBS=-lpthread
make
make install

Best regards,
Benjamin



kelvin-2 wrote:
> 
> I have tried recompiling numerous times but the results have been
> consistenly unsatisfying.  It should be noted that I am able to
> successfully install and plot on a 64 bit machine with RHEL4.  Here are
> the error messages - does anyone know where to look or what might be
> preventing the tool from working with gnuplot?
> 
> using octave-3.2.2 with gnuplot 4.3.0.
> 
> bones:octave-3.2.2 >run-octave
> warning: lo_ieee_init: unrecognized floating point format!
> 
> ...
> 
> octave:1> plot(1:10)
> error: get: invalid handle (= 1.79769e+308)
> error: called from:
> error:  
> /z/users/kmccollough/projects/octave-3.2.2/scripts/plot/__plt_get_axis_arg__.m
> at line 57, column 9
> error:   /z/users/kmccollough/projects/octave-3.2.2/scripts/plot/plot.m at
> line 178, column 18
> octave:1> __gnuplot_version__
> ans = 4.3.0
> octave:2>
> 
> Thanks,
> kelvin
> 
> 

-- 
View this message in context: http://www.nabble.com/gnuplot-not-working-in-RHEL4-32-bit-tp25224366p25347537.html
Sent from the Octave - General mailing list archive at Nabble.com.


From gbohanno at netscape.net  Mon Sep  7 23:20:37 2009
From: gbohanno at netscape.net (George Bohannon)
Date: Tue, 08 Sep 2009 00:20:37 -0400
Subject: Trouble building 3.2.2 (pthread)
In-Reply-To: <8CBFE2D75F48880-28C0-1D9C2@webmail-m034.sysops.aol.com>
References: <8CBFE2D75F48880-28C0-1D9C2@webmail-m034.sysops.aol.com>
Message-ID: <8CBFE52DA07212F-40FC-D624@webmail-d012.sysops.aol.com>












> I'm trying to build 3.2.2 on a Red Hat EL4 system. I installed gcc 4.3.4 specifically for this purpose. I ran configure successfully as follows:





> ./configure F77=gfortran





> I then ran 'make' and got the following error after building for a while. ...


I got past the pthread problem by defining the following environment variable and rerunning configure:



$ export LDFLAGS=-lpthread



However, I then ran make and encountered the following error. Any suggestions?




making doc-files
DOC_FILES = __contourc__.df __convn__.df __delaunayn__.df __dsearchn__.df __glpk__.df __lin_interpn__.df __magick_read__.
df __pchip_deriv__.df __qp__.df __voronoi__.df amd.df balance.df besselj.df betainc.df bitfcns.df bsxfun.df ccolamd.df ce
llfun.df chol.df colamd.df colloc.df conv2.df convhulln.df daspk.df dasrt.df dassl.df data.df debug.df defaults.df det.df
?dirfns.df dispatch.df dlmread.df dmperm.df eig.df eigs.df error.df fft.df fft2.df fftn.df fftw.df file-io.df filter.df f
ind.df fltk_backend.df gammainc.df gcd.df getgrent.df getpwent.df getrusage.df givens.df graphics.df help.df hess.df hex2
num.df input.df inv.df kron.df lex.df load-path.df load-save.df lookup.df ls-oct-ascii.df lsode.df lu.df luinc.df mappers
.df matrix_type.df max.df md5sum.df oct-hist.df octave.df ov-base.df ov-cell.df ov-class.df ov-fcn-handle.df ov-fcn-inlin
e.df ov-flt-re-mat.df ov-int16.df ov-int32.df ov-int64.df ov-int8.df ov-list.df ov-null-mat.df ov-re-mat.df ov-struct.df 
ov-typeinfo.df ov-uint16.df ov-uint32.df ov-uint64.df ov-uint8.df ov-usr-fcn.df ov.df pager.df parse.df pinv.df pr-output
.df pt-arg-list.df pt-eval.df pt-mat.df qr.df quad.df qz.df rand.df rcond.df regexp.df schur.df sighandlers.df sparse.df 
spparms.df sqrtm.df strfns.df svd.df syl.df symbfact.df symrcm.df symtab.df syscalls.df sysdep.df time.df toplev.df tsear
ch.df typecast.df urlwrite.df utils.df variables.df
mv doc-files-t doc-files
making gendoc.cc
g++ -g -O2 -o gendoc gendoc.cc -lpthread
making DOCSTRINGS
./gendoc: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.9' not found (required by ./gendoc)
make[2]: *** [DOCSTRINGS] Error 1
make[2]: Leaving directory `/usr/local/octave-3.2.2/src'
make[1]: *** [src] Error 2
make[1]: Leaving directory `/usr/local/octave-3.2.2'
make: *** [all] Error 2































 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090908/4edfadf8/attachment-0001.html 

From dasergatskov at gmail.com  Tue Sep  8 10:44:00 2009
From: dasergatskov at gmail.com (Dmitri A. Sergatskov)
Date: Tue, 8 Sep 2009 10:44:00 -0500
Subject: gnuplot not working in RHEL4 32 bit
In-Reply-To: <25347537.post@talk.nabble.com>
References: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
	<25347537.post@talk.nabble.com>
Message-ID: <892b16670909080844q5b83868dqe96173fb721dd387@mail.gmail.com>

>> warning: lo_ieee_init: unrecognized floating point format!

The problem is with lapack (and atlas if it builds against it).
You may want to have a look at

 https://bugzilla.redhat.com/show_bug.cgi?id=520518

and fill a similar bug report for RHEL.

Sincerely,

Dmitri.
--

From jwe at octave.org  Tue Sep  8 10:58:10 2009
From: jwe at octave.org (John W. Eaton)
Date: Tue, 8 Sep 2009 11:58:10 -0400
Subject: gnuplot not working in RHEL4 32 bit
In-Reply-To: <892b16670909080844q5b83868dqe96173fb721dd387@mail.gmail.com>
References: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
	<25347537.post@talk.nabble.com>
	<892b16670909080844q5b83868dqe96173fb721dd387@mail.gmail.com>
Message-ID: <19110.32530.318519.357455@segfault.lan>

On  8-Sep-2009, Dmitri A. Sergatskov wrote:

| >> warning: lo_ieee_init: unrecognized floating point format!

FWIW, in future versions of Octave (at least in releases after the
3.2.x series) this will be a fatal error, so users won't be able to
easily ignore it.

The patch is here:

  http://hg.savannah.gnu.org/hgweb/octave/rev/264fb5520973

and is fairly simple, so perhaps it should be applied to the 3.2.x
series as well.

jwe

From xdong.cn at gmail.com  Tue Sep  8 19:28:28 2009
From: xdong.cn at gmail.com (Xin Dong)
Date: Tue, 8 Sep 2009 19:28:28 -0500
Subject: Install packages in Mac OS X
Message-ID: <FBB7EF42-090C-4C06-94A7-860A654CA720@gmail.com>

Hi,

Has anyone successfully installed octave-forge packages with snow  
leopard and octave 3.2.2? I tried to install image-1.0.10.tar.gz, but  
failed.

Part of the error message:

   "intNDArray<octave_int<unsigned char> >::increment_index 
(Array<int>&, dim_vector const&, int)", referenced from:
       octave_value_list do_filtering<intNDArray<octave_int<unsigned  
char> >, octave_int<unsigned char>, intNDArray<octave_int<unsigned  
char> >, octave_int<unsigned char> >(intNDArray<octave_int<unsigned  
char> > const&, boolNDArray const&, octave_int<unsigned char> (*) 
(intNDArray<octave_int<unsigned char> >&, int, int),  
intNDArray<octave_int<unsigned char> > const&, int)in  
__spatial_filtering__.o
ld: symbol(s) not found
collect2: ld returned 1 exit status
make: *** [__spatial_filtering__.oct] Error 1

Thanks in advance for any help,
Xin

From tmacchant at yahoo.co.jp  Tue Sep  8 20:40:05 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Wed, 9 Sep 2009 10:40:05 +0900 (JST)
Subject: Can I change maker of stem?
In-Reply-To: <4AA6969A.9010101@dbateman.org>
Message-ID: <20090909014005.22709.qmail@web3804.mail.bbt.yahoo.co.jp>

Hello

--- David Bateman <dbateman at dbateman.org> wrote:

> Try
> 
> h = stem(1:10, 'marker', 'x')
> set (h, 'marker', 'o')
> 
> in octave 3.2.x ... This uses the stem series objects supported in 
> Octave 3.2.x and described in the section
> 
> Plotting -> Advanced Plotting -> Object Groups -> Stem Series
> 
> of the manual

h=stem(1:10, 'marker', 'x'); gives
> h=stem(1:10, 'marker', 'x');
error: subscript indices must be either positive integers or logicals.
error: called from:
error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\__add_datasource__.m at line 42,
column 5
error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\__stem__.m at line 159, column 5
error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\stem.m at line 78, column 7
octave-3.2.2.exe:15:C:\Programs\Octave\3.2.2_gcc-4.3.0\bin

Therefore I showed an example to the questioner as,

x = 1:10; 
y = ones (1, length (x))*2.*x; 
h = stem(x,y); 
set(h, 'marker','x'); 
set(h, 'marker','*'); 

Anyway thanks a lot!

Tatsuro

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From febty82 at gmail.com  Tue Sep  8 21:14:30 2009
From: febty82 at gmail.com (febty febriani)
Date: Wed, 9 Sep 2009 11:14:30 +0900
Subject: how to change scale display in imagesc
In-Reply-To: <94A29D38-8E28-4F2F-B48D-31CC7BE3A71B@illusonic.com>
References: <1964ddb30909062125o2378fa17r84dcbe971ec94fd7@mail.gmail.com>
	<94A29D38-8E28-4F2F-B48D-31CC7BE3A71B@illusonic.com>
Message-ID: <1964ddb30909081914u5a2bd99eq824fe29caf105eb7@mail.gmail.com>

thanks for your response. but it did not work for my case. I used this
command to reverse y-axis : set(gca,"ydir","normal").

your sincerely,
febty

2009/9/8 Christophe Tournery <christophe.tournery at illusonic.com>

> You may want to try:
> axis xy
> -c
>
>
> On Sep 7, 2009, at 6:25 AM, febty febriani wrote:
>
>  Dear all,
>>
>> I plotted the attached figure with using imagesc. My question is how to
>> change the scale display so that the y- scale from 0 to 0.5, not from 0.5 to
>> 0 like the attached figure. Any thoughts are appreciated. Thanks very much..
>>
>>
>> Your sincerely,
>> Febty
>> --
>>
>> ******
>> febty febriani
>> Indonesian Institute of Sciences
>> Research Center for Physics
>> Kompleks PUSPIPTEK
>> Serpong, Indonesia, 15314
>> <output20090101.png>_______________________________________________
>> Help-octave mailing list
>> Help-octave at octave.org
>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>>
>
>


-- 

******
febty febriani
Indonesian Institute of Sciences
Research Center for Physics
Kompleks PUSPIPTEK
Serpong, Indonesia, 15314
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090909/677e735a/attachment.html 

From riccardocorradini at yahoo.it  Wed Sep  9 04:27:49 2009
From: riccardocorradini at yahoo.it (Riccardo Corradini)
Date: Wed, 9 Sep 2009 09:27:49 +0000 (GMT)
Subject: R: Trouble building 3.2.2 (pthread)
In-Reply-To: <8CBFE52DA07212F-40FC-D624@webmail-d012.sysops.aol.com>
Message-ID: <287860.37973.qm@web25508.mail.ukl.yahoo.com>

Hi George,
try to install the glib rpm package for redhat
See
http://www.linuxforums.org/forum/redhat-fedora-linux-help/28627-installing-glib.html

Bests
Riccardo Corradini

--- Mar 8/9/09, George Bohannon <gbohanno at netscape.net> ha scritto:

Da: George Bohannon <gbohanno at netscape.net>
Oggetto: Trouble building 3.2.2 (pthread)
A: help-octave at octave.org
Data: Marted? 8 settembre 2009, 06:20













> I'm trying to build 3.2.2 on a Red Hat EL4 system. I installed gcc 4.3.4 specifically for this purpose. I ran configure successfully as follows:







> ./configure F77=gfortran







> I then ran 'make' and got the following error after building for a while. ...




I got past the pthread problem by defining the following environment variable and rerunning configure:





$ export LDFLAGS=-lpthread





However, I then ran make and encountered the following error. Any suggestions?








making doc-files

DOC_FILES = __contourc__.df __convn__.df __delaunayn__.df __dsearchn__.df __glpk__.df __lin_interpn__.df __magick_read__.

df __pchip_deriv__.df __qp__.df __voronoi__.df amd.df balance.df besselj.df betainc.df bitfcns.df bsxfun.df ccolamd.df ce

llfun.df chol.df colamd.df colloc.df conv2.df convhulln.df daspk.df dasrt.df dassl.df data.df debug.df defaults.df det.df

?dirfns.df dispatch.df dlmread.df dmperm.df eig.df eigs.df error.df fft.df fft2.df fftn.df fftw.df file-io.df filter.df f

ind.df fltk_backend.df gammainc.df gcd.df getgrent.df getpwent.df getrusage.df givens.df graphics.df help.df hess.df hex2

num.df input.df inv.df kron.df lex.df load-path.df load-save.df lookup.df ls-oct-ascii.df lsode.df lu.df luinc.df mappers

.df matrix_type.df max.df md5sum.df oct-hist.df octave.df ov-base.df ov-cell.df ov-class.df ov-fcn-handle.df ov-fcn-inlin

e.df ov-flt-re-mat.df ov-int16.df ov-int32.df ov-int64.df ov-int8.df ov-list.df ov-null-mat.df ov-re-mat.df ov-struct.df 

ov-typeinfo.df ov-uint16.df ov-uint32.df ov-uint64.df ov-uint8.df ov-usr-fcn.df ov.df pager.df parse.df pinv.df pr-output

.df pt-arg-list.df pt-eval.df pt-mat.df qr.df quad.df qz.df rand.df rcond.df regexp.df schur.df sighandlers.df sparse.df 

spparms.df sqrtm.df strfns.df svd.df syl.df symbfact.df symrcm.df symtab.df syscalls.df sysdep.df time.df toplev.df tsear

ch.df typecast.df urlwrite.df utils.df variables.df

mv doc-files-t doc-files

making gendoc.cc

g++ -g -O2 -o gendoc gendoc.cc -lpthread

making DOCSTRINGS

./gendoc: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.9' not found (required by ./gendoc)

make[2]: *** [DOCSTRINGS] Error 1

make[2]: Leaving directory `/usr/local/octave-3.2.2/src'

make[1]: *** [src] Error 2

make[1]: Leaving directory `/usr/local/octave-3.2.2'

make: *** [all] Error 2












































 


-----Segue allegato-----

_______________________________________________
Help-octave mailing list
Help-octave at octave.org
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090909/e97fe38f/attachment.html 

From dbateman at dbateman.org  Tue Sep  8 12:38:34 2009
From: dbateman at dbateman.org (David Bateman)
Date: Tue, 08 Sep 2009 19:38:34 +0200
Subject: Can I change maker of stem?
In-Reply-To: <20090908073942.78562.qmail@web3804.mail.bbt.yahoo.co.jp>
References: <20090908073942.78562.qmail@web3804.mail.bbt.yahoo.co.jp>
Message-ID: <4AA6969A.9010101@dbateman.org>

Tatsuro MATSUOKA wrote:
> Hello
>
> This inquiry is done in the octave thread in Japan.
>
> In the manual
>
> The default color is "r" (red). The default line style is "-" and the default marker is
> "o". The line style can be altered by the linespec argument in the same manner as
> the plot command. 
>
> However, there is no description of alternation of the marker. 
> Can I change maker of stem?  For example, "x".
>
> It will be grateful for me if someone will give me information.
>
> Regards
>
> Tatsuro  
>
> --------------------------------------
> Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
> http://pr.mail.yahoo.co.jp/ec10years/
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
>   
Try

h = stem(1:10, 'marker', 'x')
set (h, 'marker', 'o')

in octave 3.2.x ... This uses the stem series objects supported in 
Octave 3.2.x and described in the section

Plotting -> Advanced Plotting -> Object Groups -> Stem Series

of the manual

D.

-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From gkousiou at telecom.ntua.gr  Wed Sep  9 04:49:30 2009
From: gkousiou at telecom.ntua.gr (George Kousiouris)
Date: Wed, 9 Sep 2009 12:49:30 +0300 (EEST)
Subject: Command for automatic deployment of dependent packages
Message-ID: <1799.147.102.19.192.1252489770.squirrel@my.telecom.ntua.gr>


Hi all,

I have a question regarding octave installation (basically package
installation). Is there a command that adds a package to an octave
installation and automatically detects the dependencies of this package
from other packages and installs them too?

Thanks in advance and best regards,
George




From matthias.brennwald at eawag.ch  Wed Sep  9 03:44:09 2009
From: matthias.brennwald at eawag.ch (Matthias Brennwald)
Date: Wed, 9 Sep 2009 10:44:09 +0200
Subject: Exponential notation in tick labels?
Message-ID: <9156614F-3C2D-4222-93F8-D9C39C71A697@eawag.ch>

Dear all

I'm sure this has been discussed before, but I couldn't find anything  
(maybe I did not chose the right search terms). I'd like to use  
exponential notation for tick-mark labels in plots. For instance "plot  
([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,  
600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.  
Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?

Matthias

----
Matthias Brennwald
Eawag, CH-8600 D?bendorf
Office Eawag BU-C08: +41 44 823 53 05
Lab Eawag LA-C76: +41 44 823 53 06
Lab ETHZ CLA A31.6: +41 632 09 36/39
matthias.brennwald at eawag.ch



From matthias at brennwald.org  Wed Sep  9 06:18:22 2009
From: matthias at brennwald.org (Matthias Brennwald)
Date: Wed, 9 Sep 2009 13:18:22 +0200
Subject: Exponential notation in tick labels?
Message-ID: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>

Dear all

I'm sure this has been discussed before, but I couldn't find anything  
(maybe I did not chose the right search terms). I'd like to use  
exponential notation for tick-mark labels in plots. For instance "plot  
([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,  
600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.  
Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?

Matthias

----
Matthias Brennwald
Eawag, CH-8600 D?bendorf
Office Eawag BU-C08: +41 44 823 53 05
Lab Eawag LA-C76: +41 44 823 53 06
Lab ETHZ CLA A31.6: +41 632 09 36/39
matthias.brennwald at eawag.ch



From carlo.defalco at gmail.com  Wed Sep  9 07:10:26 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Wed, 9 Sep 2009 14:10:26 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
Message-ID: <420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>

2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
> Dear all
>
> I'm sure this has been discussed before, but I couldn't find anything
> (maybe I did not chose the right search terms). I'd like to use
> exponential notation for tick-mark labels in plots. For instance "plot
> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,
> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.
> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?


octave-3.2.2:40> plot ([0:1e2:1e6])
octave-3.2.2:41> xt = get(gca, 'xtick');
octave-3.2.2:42> yt = get(gca, 'ytick');
octave-3.2.2:43> set(gca, 'xticklabel', sprintf('%1.1e|', xt));
octave-3.2.2:44> set(gca, 'yticklabel', sprintf('%1.1e|', yt));


HTH,
c.

> Matthias
>
> ----
> Matthias Brennwald
> Eawag, CH-8600 D?bendorf
> Office Eawag BU-C08: +41 44 823 53 05
> Lab Eawag LA-C76: +41 44 823 53 06
> Lab ETHZ CLA A31.6: +41 632 09 36/39
> matthias.brennwald at eawag.ch
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>


From highegg at gmail.com  Wed Sep  9 07:15:55 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Wed, 9 Sep 2009 14:15:55 +0200
Subject: gnuplot not working in RHEL4 32 bit
In-Reply-To: <19110.32530.318519.357455@segfault.lan>
References: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
	<25347537.post@talk.nabble.com>
	<892b16670909080844q5b83868dqe96173fb721dd387@mail.gmail.com>
	<19110.32530.318519.357455@segfault.lan>
Message-ID: <69d8d540909090515i6cbfc5f2h8d947aa79083d5ce@mail.gmail.com>

On Tue, Sep 8, 2009 at 5:58 PM, John W. Eaton<jwe at octave.org> wrote:
> On ?8-Sep-2009, Dmitri A. Sergatskov wrote:
>
> | >> warning: lo_ieee_init: unrecognized floating point format!
>
> FWIW, in future versions of Octave (at least in releases after the
> 3.2.x series) this will be a fatal error, so users won't be able to
> easily ignore it.
>
> The patch is here:
>
> ?http://hg.savannah.gnu.org/hgweb/octave/rev/264fb5520973
>
> and is fairly simple, so perhaps it should be applied to the 3.2.x
> series as well.
>

applied.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From matthias at brennwald.org  Wed Sep  9 07:35:29 2009
From: matthias at brennwald.org (Matthias Brennwald)
Date: Wed, 9 Sep 2009 14:35:29 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
	<420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
Message-ID: <40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>


On Sep 9, 2009, at 2:10 PM, Carlo de Falco wrote:

> 2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
>> Dear all
>>
>> I'm sure this has been discussed before, but I couldn't find anything
>> (maybe I did not chose the right search terms). I'd like to use
>> exponential notation for tick-mark labels in plots. For instance  
>> "plot
>> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,
>> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5,  
>> 1e6.
>> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?
>
>
> octave-3.2.2:40> plot ([0:1e2:1e6])
> octave-3.2.2:41> xt = get(gca, 'xtick');
> octave-3.2.2:42> yt = get(gca, 'ytick');
> octave-3.2.2:43> set(gca, 'xticklabel', sprintf('%1.1e|', xt));
> octave-3.2.2:44> set(gca, 'yticklabel', sprintf('%1.1e|', yt));
>
>
> HTH,
> c.

Thanks, this does the trick! Now I could even try to adapt this to the  
"10^n" notation... I don't know a direct way to use sprintf(...) to  
print numbers in  the 10^n format, but I guess it is feasible by  
manipulating the ticklabels using string functions somehow.

Thanks again
Matthias

From dbateman at dbateman.org  Wed Sep  9 07:48:02 2009
From: dbateman at dbateman.org (David Bateman)
Date: Wed, 09 Sep 2009 14:48:02 +0200
Subject: Can I change maker of stem?
In-Reply-To: <20090909014005.22709.qmail@web3804.mail.bbt.yahoo.co.jp>
References: <20090909014005.22709.qmail@web3804.mail.bbt.yahoo.co.jp>
Message-ID: <4AA7A402.7040605@dbateman.org>

Tatsuro MATSUOKA wrote:
> Hello
>
> --- David Bateman <dbateman at dbateman.org> wrote:
>
>   
>> Try
>>
>> h = stem(1:10, 'marker', 'x')
>> set (h, 'marker', 'o')
>>
>> in octave 3.2.x ... This uses the stem series objects supported in 
>> Octave 3.2.x and described in the section
>>
>> Plotting -> Advanced Plotting -> Object Groups -> Stem Series
>>
>> of the manual
>>     
>
> h=stem(1:10, 'marker', 'x'); gives
>   
>> h=stem(1:10, 'marker', 'x');
>>     
> error: subscript indices must be either positive integers or logicals.
> error: called from:
> error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\__add_datasource__.m at line 42,
> column 5
> error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\__stem__.m at line 159, column 5
> error:   C:\Programs\Octave\3.2.2_gcc-4.3.0\share\octave\3.2.2\m\plot\stem.m at line 78, column 7
> octave-3.2.2.exe:15:C:\Programs\Octave\3.2.2_gcc-4.3.0\bin
>
>   
Oppps, check the bugfix

http://hg.tw-math.de/release-3-2-x/rev/d208ae6e9d74

so the first syntax will be fixed in 3.2.3

D.


-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From carlo.defalco at gmail.com  Wed Sep  9 07:59:04 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Wed, 9 Sep 2009 14:59:04 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
	<420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
	<40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
Message-ID: <420ba0760909090559k49dcbe80l64db6aa58cc1628c@mail.gmail.com>

2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
> Thanks, this does the trick! Now I could even try to adapt this to the
> "10^n" notation... I don't know a direct way to use sprintf(...) to
> print numbers in ?the 10^n format, but I guess it is feasible by
> manipulating the ticklabels using string functions somehow.


octave-3.2.2:76> x= [1e5:1e5:1e6];
octave-3.2.2:77> plot(x,x);
octave-3.2.2:78> axis tight
octave-3.2.2:79> xt = get(gca, 'xtick');
octave-3.2.2:80> x_ex = floor(log10(xt));
octave-3.2.2:81> x_mant = xt./10.^x_ex;
octave-3.2.2:82> set(gca, 'xticklabel', sprintf('%1.1fx10^{%d}|',
x_mant, x_ex));


Note that if one of the ticks were labeled "0" it would need to be
handled separately
(that's the reason for "axis tight")


> Thanks again
> Matthias
c.


From bpabbott at mac.com  Wed Sep  9 08:05:20 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Wed, 09 Sep 2009 09:05:20 -0400
Subject: Exponential notation in tick labels?
In-Reply-To: <40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
	<420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
	<40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
Message-ID: <110624671733611579087806112242133143711-Webmail@me.com>

On Wednesday, September 09, 2009, at 08:35AM, "Matthias Brennwald" <matthias at brennwald.org> wrote:
>
>On Sep 9, 2009, at 2:10 PM, Carlo de Falco wrote:
>
>> 2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
>>> Dear all
>>>
>>> I'm sure this has been discussed before, but I couldn't find anything
>>> (maybe I did not chose the right search terms). I'd like to use
>>> exponential notation for tick-mark labels in plots. For instance  
>>> "plot
>>> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,
>>> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5,  
>>> 1e6.
>>> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?
>>
>>
>> octave-3.2.2:40> plot ([0:1e2:1e6])
>> octave-3.2.2:41> xt = get(gca, 'xtick');
>> octave-3.2.2:42> yt = get(gca, 'ytick');
>> octave-3.2.2:43> set(gca, 'xticklabel', sprintf('%1.1e|', xt));
>> octave-3.2.2:44> set(gca, 'yticklabel', sprintf('%1.1e|', yt));
>>
>>
>> HTH,
>> c.
>
>Thanks, this does the trick! Now I could even try to adapt this to the  
>"10^n" notation... I don't know a direct way to use sprintf(...) to  
>print numbers in  the 10^n format, but I guess it is feasible by  
>manipulating the ticklabels using string functions somehow.
>
>Thanks again
>Matthias

If you're looking to fix a single figure ...

yt = '0|2\cdot{10^5}|4\cdot{10^5}|6\cdot{10^5}|8\cdot{10^5}|10^6';
set (gca, "yticklabel", yt)

Ben




From carlo.defalco at gmail.com  Wed Sep  9 08:07:00 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Wed, 9 Sep 2009 15:07:00 +0200
Subject: Command for automatic deployment of dependent packages
In-Reply-To: <1799.147.102.19.192.1252489770.squirrel@my.telecom.ntua.gr>
References: <1799.147.102.19.192.1252489770.squirrel@my.telecom.ntua.gr>
Message-ID: <420ba0760909090607y41d19b0bi6d25cd2555ad9099@mail.gmail.com>

2009/9/9 George Kousiouris <gkousiou at telecom.ntua.gr>:
>
> Hi all,
>
> I have a question regarding octave installation (basically package
> installation). Is there a command that adds a package to an octave
> installation and automatically detects the dependencies of this package
> from other packages and installs them too?

This is not currently possible as it would require the package manager
to download
the packages from the repository automatically.
On the other hand, in some linux distributions (at least fedora and
debian, AFAIK) octave packages are distributed via the system's
package manager which does usually support this feature so doing

apt-get install <package-name>

on debian or

yum install <package-name>

on fedora should do exactly what you ask

> Thanks in advance and best regards,
> George
>
c.

From carlo.defalco at gmail.com  Wed Sep  9 08:38:16 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Wed, 9 Sep 2009 15:38:16 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <420ba0760909090559k49dcbe80l64db6aa58cc1628c@mail.gmail.com>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
	<420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
	<40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
	<420ba0760909090559k49dcbe80l64db6aa58cc1628c@mail.gmail.com>
Message-ID: <420ba0760909090638k49d59660td8ea6d4c38cb4214@mail.gmail.com>

2009/9/9 Carlo de Falco <carlo.defalco at gmail.com>:
> 2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
>> Thanks, this does the trick! Now I could even try to adapt this to the
>> "10^n" notation... I don't know a direct way to use sprintf(...) to
>> print numbers in ?the 10^n format, but I guess it is feasible by
>> manipulating the ticklabels using string functions somehow.
>
>
> octave-3.2.2:76> x= [1e5:1e5:1e6];
> octave-3.2.2:77> plot(x,x);
> octave-3.2.2:78> axis tight
> octave-3.2.2:79> xt = get(gca, 'xtick');
> octave-3.2.2:80> x_ex = floor(log10(xt));
> octave-3.2.2:81> x_mant = xt./10.^x_ex;
> octave-3.2.2:82> set(gca, 'xticklabel', sprintf('%1.1fx10^{%d}|',
> x_mant, x_ex));
>
>
> Note that if one of the ticks were labeled "0" it would need to be
> handled separately
> (that's the reason for "axis tight")

sorry, I made  mistake in the call to sprintf above..
the approach below should work correctly and also andle the
zeros automatically...

function [ex, mant] = exp_notation (x)
 nz = find(x);
 ex(nz) = floor(log10(abs(x(nz))));
 mant(nz) = x(nz)./10.^ex(nz);
endfunction

octave-3.2.2:108> x= [-1e6:1e5:1e6];
octave-3.2.2:109> plot(x,x);
octave-3.2.2:110> xt = get(gca, 'xtick');[xex,xmant] = exp_notation (xt);
octave-3.2.2:111> yt = get(gca, 'ytick');[yex,ymant] = exp_notation (yt);
octave-3.2.2:112> set(gca, 'xticklabel', sprintf('%3.3gx10^{%d}|',
[xmant; xex](:)));
octave-3.2.2:113> set(gca, 'yticklabel', sprintf('%3.3gx10^{%d}|',
[ymant; yex](:)));


>> Thanks again
>> Matthias
> c.
>


From riccardocorradini at yahoo.it  Wed Sep  9 08:45:35 2009
From: riccardocorradini at yahoo.it (Riccardo Corradini)
Date: Wed, 9 Sep 2009 13:45:35 +0000 (GMT)
Subject: octave + mpi classes  into c++ code
Message-ID: <741710.47231.qm@web25502.mail.ukl.yahoo.com>

Hi I successfully installed http://www.osl.iu.edu/research/oompi/software.php
these C++ classes with open mpi 1.2.8. Are there any hints on how is possible to use them efficiently into 
or with or extending octave c++ classes .
This would be nice because? if on one hand the octave's API change and improve rapidly overt time, on the other hand it would easy to use the same mpi C++ classes.
I would like to have an example.
Thanks a lot for all your help.
Respectfully
Riccardo Corradini 




      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090909/2399de2f/attachment.html 

From matthias at brennwald.org  Wed Sep  9 09:07:00 2009
From: matthias at brennwald.org (Matthias Brennwald)
Date: Wed, 9 Sep 2009 16:07:00 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <420ba0760909090638k49d59660td8ea6d4c38cb4214@mail.gmail.com>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
	<420ba0760909090510m4b95a6cby468a93f126522b79@mail.gmail.com>
	<40BD3A73-4D21-4152-80F3-28D7D4FE418C@brennwald.org>
	<420ba0760909090559k49dcbe80l64db6aa58cc1628c@mail.gmail.com>
	<420ba0760909090638k49d59660td8ea6d4c38cb4214@mail.gmail.com>
Message-ID: <722D198C-D88A-446E-8815-B8F44434ED97@brennwald.org>


On Sep 9, 2009, at 3:38 PM, Carlo de Falco wrote:

> 2009/9/9 Carlo de Falco <carlo.defalco at gmail.com>:
>> 2009/9/9 Matthias Brennwald <matthias at brennwald.org>:
>>> Thanks, this does the trick! Now I could even try to adapt this to  
>>> the
>>> "10^n" notation... I don't know a direct way to use sprintf(...) to
>>> print numbers in  the 10^n format, but I guess it is feasible by
>>> manipulating the ticklabels using string functions somehow.
>>
>>
>> octave-3.2.2:76> x= [1e5:1e5:1e6];
>> octave-3.2.2:77> plot(x,x);
>> octave-3.2.2:78> axis tight
>> octave-3.2.2:79> xt = get(gca, 'xtick');
>> octave-3.2.2:80> x_ex = floor(log10(xt));
>> octave-3.2.2:81> x_mant = xt./10.^x_ex;
>> octave-3.2.2:82> set(gca, 'xticklabel', sprintf('%1.1fx10^{%d}|',
>> x_mant, x_ex));
>>
>>
>> Note that if one of the ticks were labeled "0" it would need to be
>> handled separately
>> (that's the reason for "axis tight")
>
> sorry, I made  mistake in the call to sprintf above..
> the approach below should work correctly and also andle the
> zeros automatically...
>
> function [ex, mant] = exp_notation (x)
> nz = find(x);
> ex(nz) = floor(log10(abs(x(nz))));
> mant(nz) = x(nz)./10.^ex(nz);
> endfunction
>
> octave-3.2.2:108> x= [-1e6:1e5:1e6];
> octave-3.2.2:109> plot(x,x);
> octave-3.2.2:110> xt = get(gca, 'xtick');[xex,xmant] = exp_notation  
> (xt);
> octave-3.2.2:111> yt = get(gca, 'ytick');[yex,ymant] = exp_notation  
> (yt);
> octave-3.2.2:112> set(gca, 'xticklabel', sprintf('%3.3gx10^{%d}|',
> [xmant; xex](:)));
> octave-3.2.2:113> set(gca, 'yticklabel', sprintf('%3.3gx10^{%d}|',
> [ymant; yex](:)));

Even better, thanks a bunch!

Matthias

From vic at norton.name  Wed Sep  9 09:23:55 2009
From: vic at norton.name (Vic Norton)
Date: Wed, 9 Sep 2009 10:23:55 -0400
Subject: Install packages in Mac OS X
In-Reply-To: <FBB7EF42-090C-4C06-94A7-860A654CA720@gmail.com>
References: <FBB7EF42-090C-4C06-94A7-860A654CA720@gmail.com>
Message-ID: <97579901-A788-4093-AD33-4FBDE5C1D709@norton.name>

I don't have Snow Leopard yet, but you might try this. It works for me.
    1) Download "Octave.app for Mac OS X" from
          <http://octave.sourceforge.net/>.
    2) Drop the Octave application into your Applications folder.
    3) To be able to use the shebang line
          #!/usr/local/bin/octave
       in your scripts, do
          $ sudo ln -s \
            /Applications/Octave.app/Contents/Resources/bin/octave \
            /usr/local/bin
       in Terminal.

Regards,

Vic


On Sep 8, 2009, at 8:28 PM, Xin Dong wrote:

> Hi,
>
> Has anyone successfully installed octave-forge packages with snow
> leopard and octave 3.2.2? I tried to install image-1.0.10.tar.gz, but
> failed.


From choonway at gmail.com  Wed Sep  9 11:08:22 2009
From: choonway at gmail.com (Liao Choon Way)
Date: Wed, 9 Sep 2009 09:08:22 -0700 (PDT)
Subject: unofficial octave 3.2.2 msvc2008 solution/project files
Message-ID: <25368142.post@talk.nabble.com>


hi all,

i have prepared a _rough_ set of files for those interested in embedding
octave
into windows applications and have problems with mingw dlls.

Get them 
http://sourceforge.net/projects/el-rekka/files/octave-3.2.2-msvc9-20090909.tar.gz/download
here .

instructions enclosed. read before use. check the repository for updates.

choonway
-- 
View this message in context: http://www.nabble.com/unofficial-octave-3.2.2-msvc2008-solution-project-files-tp25368142p25368142.html
Sent from the Octave - General mailing list archive at Nabble.com.


From bpabbott at mac.com  Wed Sep  9 06:57:55 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Wed, 09 Sep 2009 07:57:55 -0400
Subject: Exponential notation in tick labels?
In-Reply-To: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
References: <EBCC2195-77EA-4D01-A771-5880D2895BD1@brennwald.org>
Message-ID: <376D4B5A-9634-406D-867D-75FFBCAF61F4@mac.com>

On Sep 9, 2009, at 7:18 AM, Matthias Brennwald wrote:

> Dear all
>
> I'm sure this has been discussed before, but I couldn't find anything
> (maybe I did not chose the right search terms). I'd like to use
> exponential notation for tick-mark labels in plots. For instance "plot
> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,
> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.
> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?
>
> Matthias


You can over-ride the automatic formatting by ...

	set (gca, "yticklabel", num2str (get (gca, "ytick"), '%.1e|'))

Ben


From Thomas.Treichl at gmx.net  Wed Sep  9 13:15:50 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Wed, 09 Sep 2009 20:15:50 +0200
Subject: Install packages in Mac OS X
In-Reply-To: <97579901-A788-4093-AD33-4FBDE5C1D709@norton.name>
References: <FBB7EF42-090C-4C06-94A7-860A654CA720@gmail.com>
	<97579901-A788-4093-AD33-4FBDE5C1D709@norton.name>
Message-ID: <4AA7F0D6.9060206@gmx.net>

Vic Norton schrieb:
> I don't have Snow Leopard yet, but you might try this. It works for me.
>     1) Download "Octave.app for Mac OS X" from
>           <http://octave.sourceforge.net/>.
>     2) Drop the Octave application into your Applications folder.
>     3) To be able to use the shebang line
>           #!/usr/local/bin/octave
>        in your scripts, do
>           $ sudo ln -s \
>             /Applications/Octave.app/Contents/Resources/bin/octave \
>             /usr/local/bin
>        in Terminal.
> 
> Regards,
> 
> Vic
> 
> On Sep 8, 2009, at 8:28 PM, Xin Dong wrote:
>> Hi,
>>
>> Has anyone successfully installed octave-forge packages with snow
>> leopard and octave 3.2.2? I tried to install image-1.0.10.tar.gz, but
>> failed.

Hi Vic, hi Xin,

I think the problem is a bit more complex. Apple has changed some more things 
from Leopard to Snow Leopard and with Snow Leopard everything that has to do 
with Octave's mkoctfile is broken now (ie. installing packages, the 'mkoctfile' 
command itself, the command 'mex', and so on) - in detail: Snow Leopard's linker 
does not accept my 10.4 build libraries anymore and that's why linking against 
my binaries completely fails. I've already commented on a similar problem here

   http://www-old.cae.wisc.edu/pipermail/bug-octave/2009-September/009375.html

Just because of my own interest - what is the current GCC version from Snow 
Leopard's XCode tools? Is there at least a gfortran available?

Best regards

   Thomas



From gbohanno at netscape.net  Wed Sep  9 13:26:56 2009
From: gbohanno at netscape.net (George Bohannon)
Date: Wed, 09 Sep 2009 14:26:56 -0400
Subject: Trouble building 3.2.2 (pthread)
Message-ID: <8CBFF923E5A74CF-4584-20E4@webmail-m006.sysops.aol.com>


Here is a little more information:

There are two versions of gcc
installed on this RHEL4 system. Octave 3.2.2 will not build with the
gcc version already installed (3.4.6), so I
installed gcc 4.3.4.? The rather long output from 'make' prior to the gendoc error indicates that 'make' is using the gcc 4.3.4 library for everything up to the point at which it stops on gendoc. 

Nevertheless, here is information about the two installations:

[root at fluffy ~]# strings /usr/lib/libstdc++.so.6 | grep GLIBCXX
GLIBCXX_3.4
GLIBCXX_3.4.1
GLIBCXX_3.4.2
GLIBCXX_3.4.3
GLIBCXX_FORCE_NEW


[root at fluffy ~]# strings /usr/local/lib/libstdc++.so.6 | grep GLIBCXX
GLIBCXX_3.4
GLIBCXX_3.4.1
GLIBCXX_3.4.2
GLIBCXX_3.4.3
GLIBCXX_3.4.4
GLIBCXX_3.4.5
GLIBCXX_3.4.6
GLIBCXX_3.4.7
GLIBCXX_3.4.8
GLIBCXX_3.4.9
GLIBCXX_3.4.10
GLIBCXX_FORCE_NEW
GLIBCXX_DEBUG_MESSAGE_LENGTH


So it appears that the gcc in /usr/local/lib (which is gcc 4.3.4) has the required library. However, I have tried setting the following environment and then running configure & make without success:

export F77=gfortran
export LIBS=-lpthread
export LDFLAGS=-L/usr/local/lib

This still results in the message 
./gendoc: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.9' not found (required by ./gendoc)

I have run out of things to try. Any suggestions? 

?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090909/df42bde0/attachment.html 

From bpabbott at mac.com  Wed Sep  9 14:02:40 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Wed, 09 Sep 2009 15:02:40 -0400
Subject: Install packages in Mac OS X
Message-ID: <23337479393450908730213684685666271541-Webmail@me.com>

On Wednesday, September 09, 2009, at 02:15PM, "Thomas Treichl" <Thomas.Treichl at gmx.net> wrote:
>Vic Norton schrieb:
>> I don't have Snow Leopard yet, but you might try this. It works for me.
>>     1) Download "Octave.app for Mac OS X" from
>>           <http://octave.sourceforge.net/>.
>>     2) Drop the Octave application into your Applications folder.
>>     3) To be able to use the shebang line
>>           #!/usr/local/bin/octave
>>        in your scripts, do
>>           $ sudo ln -s \
>>             /Applications/Octave.app/Contents/Resources/bin/octave \
>>             /usr/local/bin
>>        in Terminal.
>> 
>> Regards,
>> 
>> Vic
>> 
>> On Sep 8, 2009, at 8:28 PM, Xin Dong wrote:
>>> Hi,
>>>
>>> Has anyone successfully installed octave-forge packages with snow
>>> leopard and octave 3.2.2? I tried to install image-1.0.10.tar.gz, but
>>> failed.
>
>Hi Vic, hi Xin,
>
>I think the problem is a bit more complex. Apple has changed some more things 
>from Leopard to Snow Leopard and with Snow Leopard everything that has to do 
>with Octave's mkoctfile is broken now (ie. installing packages, the 'mkoctfile' 
>command itself, the command 'mex', and so on) - in detail: Snow Leopard's linker 
>does not accept my 10.4 build libraries anymore and that's why linking against 
>my binaries completely fails. I've already commented on a similar problem here
>
>   http://www-old.cae.wisc.edu/pipermail/bug-octave/2009-September/009375.html
>
>Just because of my own interest - what is the current GCC version from Snow 
>Leopard's XCode tools? Is there at least a gfortran available?
>
>Best regards
>
>   Thomas

Hi Thomas,

I've been running Snow Leopard since the morning it was released.

$ gcc --version
i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5646)

I don't think gfortran is included (it's not in /usr/bin).

The biggest annoyance has been that the old .la files no longer work (my Fink packages are currently a mess). I deleted all .la files in the Fink directory structure and rebuild the appropriate package when an .la file comes up missing.

I can *almost* build a functioning Octave. It compiles, but crashes during "make check". If I run and then exit, I get a core dump. I wish I had time to look into what is happening but I have very little spare time at present.

Ben


From rob at jlab.org  Wed Sep  9 14:44:45 2009
From: rob at jlab.org (Rob Mahurin)
Date: Wed, 9 Sep 2009 15:44:45 -0400
Subject: Logarithmic hist and other colour
In-Reply-To: <200909081037.20947.schubert.seb@googlemail.com>
References: <200909081037.20947.schubert.seb@googlemail.com>
Message-ID: <20090909194405.GA64737@dyn21924529.dz.ornl.gov>

On Tue, Sep 08, 2009 at 10:37:16AM +0200, Sebastian Schubert wrote:
> is it possible to plot a histogram logarithmically within octave (3.2.2) 
> with the standard tools? Can I change the colour (no filling of the 
> bars)? Pointers to the manual are fine. :)

I do this by

	> hist(whatever);
	> set(gca,"yscale","log")

Cheers,
Rob

From efrios at yahoo.com  Wed Sep  9 16:28:17 2009
From: efrios at yahoo.com (Hielos)
Date: Wed, 9 Sep 2009 14:28:17 -0700 (PDT)
Subject: jhandles with debian
In-Reply-To: <6ec71d090905231528u6265c95fw2db40e30681e894a@mail.gmail.com>
References: <6ec71d090905210458l3d18fe82r870a7b8c341419ba@mail.gmail.com>
	<128f38bd0905210545s40a3fcb3iab142a26964b095a@mail.gmail.com>
	<6ec71d090905210634i31dbc3c6yfa8407cad5ae1ada@mail.gmail.com>
	<128f38bd0905210749t73e5d6f0re2dab67570e543d@mail.gmail.com>
	<6ec71d090905210756p4cda351bhfe4f822486eb454e@mail.gmail.com>
	<128f38bd0905210802k35028c58qe3278fca58a55fca@mail.gmail.com>
	<6ec71d090905231528u6265c95fw2db40e30681e894a@mail.gmail.com>
Message-ID: <25373464.post@talk.nabble.com>


Hello. If it makes you feel better, I am having the same issue as you. There
is not much information on how to solve it, and I think it should. May be,
jhandles was only destined to work with Windows..? As a matter of fact, I
think that the bottom line goes to installing octave-java. I am running
Jaunty (ubuntu 9.04) 64 bits. From my perspective, this is a REAL issue,
since I can't program any GUI's without jhandles (I have seen it's marked as
whish-list somewhere else). Good luck to all of us



oz123 wrote:
> 
> Hello, again
> ./configure on my debian machine will succeed only if I do:
> 
> export JAVA_HOME="/usr/lib/jvm/java-6-sun-1.6.0.13/"
> 
> These are the last line of the configure out put.
> 
> But like I said before it fails on make .
> __java__.cc: In member function ?octave_value
> octave_java::do_java_set(JNIEnv*, const std::string&, const
> octave_value&)?:
> __java__.cc:1812: warning: deprecated conversion from string constant to
> ?char*?
> __java__.cc: In static member function ?static octave_value
> octave_java::do_java_set(JNIEnv*, const std::string&, const std::string&,
> const octave_value&)?:
> __java__.cc:1834: warning: deprecated conversion from string constant to
> ?char*?
> make[1]: *** [__java__.oct] Error 1
> make[1]: Leaving directory `/home/ozdeb/Desktop/java-1.2.6/src'
> make: *** [build] Error 2
> 
> Are these the header files you mean ?
> 
> /usr/lib/jvm/java-6-sun-1.6.0.13/include
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/linux
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/linux/jni_md.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/linux/jawt_md.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/jni.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/jdwpTransport.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/jawt.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/classfile_constants.h
> /usr/lib/jvm/java-6-sun-1.6.0.13/include/jvmti.h
> 
> 
> Because if so, they are already under java home.
> 
> I even tried copying them to the working directory of java/src/ where make
> is working. But it still fails.
> 
> Any idea ?
> B.T.W the silence about the subject means nobody is using jhandles on
> linux
> ? Someone ?
> 
> Thanks, Oz
> 
> 
> On Thu, May 21, 2009 at 5:02 PM, Michael Goffioul <
> michael.goffioul at gmail.com> wrote:
> 
>> On Thu, May 21, 2009 at 3:56 PM, Oz Nahum <nahumoz at gmail.com> wrote:
>> >>JDK include directory is not in the standard include path, so it
>> >>seems you have to add it manually, through some environment
>> >>variable. Under MSVC, I use INCLUDE variable.
>> >
>> > Ok, could someone translate this to noobs simple language ?
>> > I understand you develop this software under microsoft windows [:-(],
>> and
>> I
>> > don't understand
>> > what is the include variable. You mean I have to change manually the
>> > configure or make files ?
>>
>> What it means is that the headers provided by JDK are not installed
>> in some standard location, so the compiler cannot find them
>> automatically.
>> The configure script expects them in $JAVA_HOME/include. If this is not
>> the case, then you'll have to fix your system, or let the compiler know
>> about
>> the non standard location, usually by using some well-known environment
>> variable. Under MSVC, this is achieved through the variable called
>> "INCLUDE".
>>
>> I don't know the equivalent for GCC, but I guess using CPPFLAGS and the
>> -I compiler flag could do it.
>>
>> Michael.
>>
> 
> 
> 
> -- 
> ----
>            Imagine there's no countries
>            It isn't hard to do
>            Nothing to kill or die for
>            And no religion too
>            Imagine all the people
>            Living life in peace
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

-- 
View this message in context: http://www.nabble.com/jhandles-with-debian-tp23652114p25373464.html
Sent from the Octave - General mailing list archive at Nabble.com.



From tmacchant at yahoo.co.jp  Wed Sep  9 19:06:56 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Thu, 10 Sep 2009 09:06:56 +0900 (JST)
Subject: Can I change maker of stem?
In-Reply-To: <4AA7A402.7040605@dbateman.org>
Message-ID: <20090910000657.93938.qmail@web3809.mail.bbt.yahoo.co.jp>

Hello

--- David Batemanwrote:

> Oppps, check the bugfix
> 
> http://hg.tw-math.de/release-3-2-x/rev/d208ae6e9d74
> 
> so the first syntax will be fixed in 3.2.3

I have  confirmed on tbe 3.2.3rc-3.

Thanks!!

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From gbohanno at netscape.net  Wed Sep  9 21:00:08 2009
From: gbohanno at netscape.net (George Bohannon)
Date: Wed, 09 Sep 2009 22:00:08 -0400
Subject: Trouble building 3.2.2 (pthread)
Message-ID: <8CBFFD18EB0F5CD-25C0-6700@webmail-m036.sysops.aol.com>

Success!? Persistence pays off again.

Gotta use lib64, not lib. Here are the environment variables that allowed the build to work:

export F77=gfortran
export LIBS=-lpthread
export LDFLAGS=-L/usr/local/lib64
export LD_LIBRARY_PATH=/usr/local/lib64

It all seems so simple now.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090909/6168a317/attachment-0001.html 

From udippel at uniten.edu.my  Wed Sep  9 21:27:13 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Thu, 10 Sep 2009 10:27:13 +0800
Subject: Octave and Eidors
Message-ID: <4AA86401.8080502@uniten.edu.my>

Dear all,

I am a total newbie. Not to Unix nor Linux, but to Octave and Eidors. I
followed what is written on the main site of eidors, but the examples fail:

octave:1> run downloads/eidors-v3.3/startup.m
warning: isstr is obsolete and will be removed from a future version of
Octave, please use ischar instead
EIDORS:[Completed setting up of EIDORS Version 3.3 ($Date: 2008/07/19
16:14:18 $)]
EIDORS:[Parameter: cache_size=100 MB]
EIDORS:[Parameter: mapped_colour=127]
EIDORS:[Default background colour: write]
EIDORS:[EIDORS mex folder:
/home/udippel/downloads/eidors-v3.3/arch/octave/pc]
octave:3> run downloads/eidors-v3.3/examples/demo_real.m
step 1: create FEM model structure
step 2: create FEM model electrodes definitions
step 3: create FEM model stimulation and measurement patterns
step 4: simulate data for homogeneous medium
step 5: simulate data for inhomogeneous medium
warning: set: invalid property `Name'
warning: set: invalid property `EdgeColor'
error: invalid value = Direct
error: called from `__patch__' in file
/usr/share/octave/3.0.1/m/plot/__patch__.m near line 182, column 5
error: gnuplot (as of v4.2) only supports 2D filled patches
error: evaluating if command near line 537, column 9
error: evaluating if command near line 531, column 7
error: evaluating if command near line 529, column 5
error: evaluating for command near line 518, column 3
error: evaluating switch command near line 332, column 7
error: evaluating for command near line 328, column 5
error: evaluating if command near line 25, column 3
error: called from `__go_draw_axes__' in file
/usr/share/octave/3.0.1/m/plot/__go_draw_axes__.m
error: evaluating switch command near line 58, column 4
error: evaluating for command near line 56, column 2
error: evaluating if command near line 33, column 7
error: evaluating if command near line 26, column 5
error: evaluating if command near line 25, column 3
error: called from `__go_draw_figure__' in file
/usr/share/octave/3.0.1/m/plot/__go_draw_figure__.m
error: evaluating if command near line 81, column 6
error: evaluating if command near line 78, column 4
error: evaluating if command near line 76, column 2
error: evaluating for command near line 75, column 7
error: evaluating if command near line 45, column 5
error: called from `drawnow' in file
/usr/share/octave/3.0.1/m/plot/drawnow.m
octave:3>

Please, have a bit of patience with me, and help me out; what is wrong??

(I have done my homework, as far as the first error is concerned, and
googled for it, but it did not result in anything. Meaning, I must be
doing something totally wrong?

Uwe


From jwe at octave.org  Thu Sep 10 00:19:17 2009
From: jwe at octave.org (John W. Eaton)
Date: Thu, 10 Sep 2009 01:19:17 -0400
Subject: Octave and Eidors
In-Reply-To: <4AA86401.8080502@uniten.edu.my>
References: <4AA86401.8080502@uniten.edu.my>
Message-ID: <19112.35925.323307.343873@segfault.lan>

On 10-Sep-2009, Uwe Dippel wrote:

| I am a total newbie. Not to Unix nor Linux, but to Octave and Eidors. I
| followed what is written on the main site of eidors, but the examples fail:
| 
| octave:1> run downloads/eidors-v3.3/startup.m
| warning: isstr is obsolete and will be removed from a future version of
| Octave, please use ischar instead
| EIDORS:[Completed setting up of EIDORS Version 3.3 ($Date: 2008/07/19
| 16:14:18 $)]
| EIDORS:[Parameter: cache_size=100 MB]
| EIDORS:[Parameter: mapped_colour=127]
| EIDORS:[Default background colour: write]
| EIDORS:[EIDORS mex folder:
| /home/udippel/downloads/eidors-v3.3/arch/octave/pc]
| octave:3> run downloads/eidors-v3.3/examples/demo_real.m
| step 1: create FEM model structure
| step 2: create FEM model electrodes definitions
| step 3: create FEM model stimulation and measurement patterns
| step 4: simulate data for homogeneous medium
| step 5: simulate data for inhomogeneous medium
| warning: set: invalid property `Name'
| warning: set: invalid property `EdgeColor'
| error: invalid value = Direct
| error: called from `__patch__' in file
| /usr/share/octave/3.0.1/m/plot/__patch__.m near line 182, column 5

You might have better luck with graphics if you try the current Octave
release, 3.2.2.

jwe

From dbateman at dbateman.org  Thu Sep 10 00:34:28 2009
From: dbateman at dbateman.org (David Bateman)
Date: Thu, 10 Sep 2009 07:34:28 +0200
Subject: Octave and Eidors
In-Reply-To: <4AA86401.8080502@uniten.edu.my>
References: <4AA86401.8080502@uniten.edu.my>
Message-ID: <4AA88FE4.4050408@dbateman.org>

Uwe Dippel wrote:
> Dear all,
>
> I am a total newbie. Not to Unix nor Linux, but to Octave and Eidors. I
> followed what is written on the main site of eidors, but the examples fail:
>
> octave:1> run downloads/eidors-v3.3/startup.m
> warning: isstr is obsolete and will be removed from a future version of
> Octave, please use ischar instead
> EIDORS:[Completed setting up of EIDORS Version 3.3 ($Date: 2008/07/19
> 16:14:18 $)]
> EIDORS:[Parameter: cache_size=100 MB]
> EIDORS:[Parameter: mapped_colour=127]
> EIDORS:[Default background colour: write]
> EIDORS:[EIDORS mex folder:
> /home/udippel/downloads/eidors-v3.3/arch/octave/pc]
> octave:3> run downloads/eidors-v3.3/examples/demo_real.m
> step 1: create FEM model structure
> step 2: create FEM model electrodes definitions
> step 3: create FEM model stimulation and measurement patterns
> step 4: simulate data for homogeneous medium
> step 5: simulate data for inhomogeneous medium
> warning: set: invalid property `Name'
> warning: set: invalid property `EdgeColor'
> error: invalid value = Direct
> error: called from `__patch__' in file
> /usr/share/octave/3.0.1/m/plot/__patch__.m near line 182, column 5
> error: gnuplot (as of v4.2) only supports 2D filled patches
> error: evaluating if command near line 537, column 9
> error: evaluating if command near line 531, column 7
> error: evaluating if command near line 529, column 5
> error: evaluating for command near line 518, column 3
> error: evaluating switch command near line 332, column 7
> error: evaluating for command near line 328, column 5
> error: evaluating if command near line 25, column 3
> error: called from `__go_draw_axes__' in file
> /usr/share/octave/3.0.1/m/plot/__go_draw_axes__.m
> error: evaluating switch command near line 58, column 4
> error: evaluating for command near line 56, column 2
> error: evaluating if command near line 33, column 7
> error: evaluating if command near line 26, column 5
> error: evaluating if command near line 25, column 3
> error: called from `__go_draw_figure__' in file
> /usr/share/octave/3.0.1/m/plot/__go_draw_figure__.m
> error: evaluating if command near line 81, column 6
> error: evaluating if command near line 78, column 4
> error: evaluating if command near line 76, column 2
> error: evaluating for command near line 75, column 7
> error: evaluating if command near line 45, column 5
> error: called from `drawnow' in file
> /usr/share/octave/3.0.1/m/plot/drawnow.m
> octave:3>
>
> Please, have a bit of patience with me, and help me out; what is wrong??
>
> (I have done my homework, as far as the first error is concerned, and
> googled for it, but it did not result in anything. Meaning, I must be
> doing something totally wrong?
>
> Uwe
>   
Try using a newer version of Octave. Version 3.2.2 for example.. I 
suspect this will work there.. Though you should report to Eidors that 
"isstr" is deprecated in Octave

David


-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From tmacchant at yahoo.co.jp  Thu Sep 10 05:14:40 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Thu, 10 Sep 2009 19:14:40 +0900 (JST)
Subject: Octave and MSVC
In-Reply-To: <981987.2481.qm@web59002.mail.re1.yahoo.com>
Message-ID: <20090910101440.23897.qmail@web3802.mail.bbt.yahoo.co.jp>

Hello

Sorry! The page is no longer maintained. I have to close the page.
The release of octave on MSVC have deleted from the sourceforge cite due to the licence issue. 

Today on the octave ML, the following topic is submitted.

unofficial octave 3.2.2 msvc2008 solution/project files
http://www.nabble.com/unofficial-octave-3.2.2-msvc2008-solution-project-files-tt25368142.html

I have not seen it yet. I hope that the above is useful for you.

Regards

Tatsuro 

--- David Sarraf  wrote:

> Dr. Matsuoka
> 
> On your "Tatsuro's Octave for Windows Page", you mention compiling .oct files with MSVC. Have
> you found a place to download the older 2005 verison of MSVC? If so, could you please provide
> me with the link? I have a similar problem. I want to use fracshift.m from the Signals
> toolbox, however that function depends on another function, upfirdn, that needs to be compiled.
> 
> 
> 
> Thank you
> 
> 
> 
> David Sarraf
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From mirek at zind.ikem.pwr.wroc.pl  Thu Sep 10 05:29:18 2009
From: mirek at zind.ikem.pwr.wroc.pl (Miroslaw Kwasniak)
Date: Thu, 10 Sep 2009 12:29:18 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <9156614F-3C2D-4222-93F8-D9C39C71A697@eawag.ch>
References: <9156614F-3C2D-4222-93F8-D9C39C71A697@eawag.ch>
Message-ID: <20090910102918.GC1544@zind.ikem.pwr.wroc.pl>

On Wed, Sep 09, 2009 at 10:44:09AM +0200, Matthias Brennwald wrote:
> Dear all
> 
> I'm sure this has been discussed before, but I couldn't find anything  
> (maybe I did not chose the right search terms). I'd like to use  
> exponential notation for tick-mark labels in plots. For instance "plot  
> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,  
> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.  
> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?

%tick values
tv=(0:2:10)*1e5;

%tick labels
tl=eval(strrep(['tl=[',sprintf("'%.0g'\n",tv),']'],'e+0','*10^'));

%use on plot
set(gca,'ytick',tv,'yticklabel',tl)

From mirek at zind.ikem.pwr.wroc.pl  Thu Sep 10 05:29:18 2009
From: mirek at zind.ikem.pwr.wroc.pl (Miroslaw Kwasniak)
Date: Thu, 10 Sep 2009 12:29:18 +0200
Subject: Exponential notation in tick labels?
In-Reply-To: <9156614F-3C2D-4222-93F8-D9C39C71A697@eawag.ch>
References: <9156614F-3C2D-4222-93F8-D9C39C71A697@eawag.ch>
Message-ID: <20090910102918.GC1544@zind.ikem.pwr.wroc.pl>

On Wed, Sep 09, 2009 at 10:44:09AM +0200, Matthias Brennwald wrote:
> Dear all
> 
> I'm sure this has been discussed before, but I couldn't find anything  
> (maybe I did not chose the right search terms). I'd like to use  
> exponential notation for tick-mark labels in plots. For instance "plot  
> ([0:1e4:1E6])" yields a plot with y-axis labels 0, 200000, 400000,  
> 600000, 700000, 1e+6. I'd like these to be 0, 2e5, 4e5, 6e5, 8e5, 1e6.  
> Or, even better, 0, 2x10^5, 4x10^5,..., 10^6. How can I achieve this?

%tick values
tv=(0:2:10)*1e5;

%tick labels
tl=eval(strrep(['tl=[',sprintf("'%.0g'\n",tv),']'],'e+0','*10^'));

%use on plot
set(gca,'ytick',tv,'yticklabel',tl)

From vic at norton.name  Thu Sep 10 05:34:59 2009
From: vic at norton.name (Vic Norton)
Date: Thu, 10 Sep 2009 06:34:59 -0400
Subject: Install packages in Mac OS X
In-Reply-To: <4AA7F0D6.9060206@gmx.net>
References: <FBB7EF42-090C-4C06-94A7-860A654CA720@gmail.com>
	<97579901-A788-4093-AD33-4FBDE5C1D709@norton.name>
	<4AA7F0D6.9060206@gmx.net>
Message-ID: <8EB1C2D9-1B3D-4DFC-AC7A-86A373C7508A@norton.name>

Thanks Thomas! This thread settles one thing for me. I have a copy of  
Snow Leopard coming from Amazon. I definitely will NOT install it  
until I can install an Octave that will work with it. I depend on  
Octave. I can do without Snow Leopard.

Regards,

Vic

On Sep 9, 2009, at 2:15 PM, Thomas Treichl wrote:

> Vic Norton schrieb:
>> I don't have Snow Leopard yet, but you might try this. It works for  
>> me.
>>    1) Download "Octave.app for Mac OS X" from
>>          <http://octave.sourceforge.net/>.
>>    2) Drop the Octave application into your Applications folder.
>>    3) To be able to use the shebang line
>>          #!/usr/local/bin/octave
>>       in your scripts, do
>>          $ sudo ln -s \
>>            /Applications/Octave.app/Contents/Resources/bin/octave \
>>            /usr/local/bin
>>       in Terminal.
>> Regards,
>> Vic
>> On Sep 8, 2009, at 8:28 PM, Xin Dong wrote:
>>> Hi,
>>>
>>> Has anyone successfully installed octave-forge packages with snow
>>> leopard and octave 3.2.2? I tried to install image-1.0.10.tar.gz,  
>>> but
>>> failed.
>
> Hi Vic, hi Xin,
>
> I think the problem is a bit more complex. Apple has changed some  
> more things from Leopard to Snow Leopard and with Snow Leopard  
> everything that has to do with Octave's mkoctfile is broken now (ie.  
> installing packages, the 'mkoctfile' command itself, the command  
> 'mex', and so on) - in detail: Snow Leopard's linker does not accept  
> my 10.4 build libraries anymore and that's why linking against my  
> binaries completely fails. I've already commented on a similar  
> problem here
>
>  http://www-old.cae.wisc.edu/pipermail/bug-octave/2009-September/009375.html
>
> Just because of my own interest - what is the current GCC version  
> from Snow Leopard's XCode tools? Is there at least a gfortran  
> available?
>
> Best regards
>
>  Thomas
>
>


From wilson.rosa at gmail.com  Thu Sep 10 09:18:52 2009
From: wilson.rosa at gmail.com (Wilson de Oliveira)
Date: Thu, 10 Sep 2009 11:18:52 -0300
Subject: Netlab on Octave
Message-ID: <91e5dab20909100718g49d98773sfc70abb694b45aad@mail.gmail.com>

Hi. I hope anyone may help me
As you may guess I am a complete newbie in Octave,
but I need to run Netlab and have no access to Matlab.

I am using Gnu Octave for Windows 3.0.0.
When trying to use the Netlab 3.3 demo ?demnlab?
I get the following messages:

octave-3.0.0.exe:11> demnlab
demnlab demnlab.m
octave-3.0.0.exe:11> demnlab
warning: set: invalid property `Name'
warning: set: invalid property `NumberTitle'
error: `uicontrol' undefined near line 33 column 3
error: evaluating if command near line 20, column 1
error: called from `demnlab' in file `c:\octave\Netlab\demnlab.m'
octave-3.0.0.exe:11>

.
Wilson de Oliveira
UFRPE-DEInfo
http://www.cin.ufpe.br/~wrdo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090910/675a0ce2/attachment.html 

From ct529 at york.ac.uk  Thu Sep 10 09:57:47 2009
From: ct529 at york.ac.uk (Corrado)
Date: Thu, 10 Sep 2009 15:57:47 +0100
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
References: <200909071338.28862.ct529@york.ac.uk>
	<200909071521.39612.ct529@york.ac.uk>
	<69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
Message-ID: <200909101557.47893.ct529@york.ac.uk>

Dear Jaroslav, Octave list,

here I am again ....

How do you calculate distance or dissimilarity matrices in Octave? and how do 
you treat triangular matrices + diagonal in Octave?  is there a separate 
package?

Regards


On Monday 07 September 2009 16:04:35 Jaroslav Hajek wrote:
> On Mon, Sep 7, 2009 at 4:21 PM, Corrado<ct529 at york.ac.uk> wrote:
> > Dear Jaroslav,
> >
> > thanks for your help.
> >
> > I actually do not understand what you are suggesting:
> >
> > 1) The data would be stored in matrices, which I can easily dump into csv
> > files
>
> OK. But mind you, instead of working with a number of equal-sized
> matrices it is often more efficient to work with a 3D array. Usually
> it depends on where you get the matrices from.
>
> > 2) The distance matrices are up to 5000 x 5000 (but they symmetric, so
> > only n(n-1)/2 are important), but the ones I am using now are just 1500 x
> > 1500.
>
> In that case, it probably doesn't matter much which representation you
> choose.
>
> > 3) I thought lsqnonneg only worked on linear system in the form y=ax
>
> Yes. That's why you need to bring the system into this form.
>
> > 4) The terms are matrices, how do you write a canonical form???? the x1
> > .... xn are matrices (e.g. 1500 x 1500) .... the ci are coefficients in R
> > ....
>
> A matrix equation constitutes a system of equations; one for each
> element. If x1...xN are MxM, you get a
> system of M*(M+1)/2 equations in N variables.
>
> Suppose you have the matrices X in a cell array, so that X{1} is the
> first matrix etc., and the RHS matrix is Y.
> Then you want to do something like
> A = [cellfun (@vech, X, "UniformOutput", false){:}];
> b = vech (Y);
> c = lsqnonneg (A, b);
> To get the coefficients in "c". Note that lsqnonneg is part of Octave
> 3.2.0+. There is a small problem with the above simplistic code: it does
> not discard the diagonal equations, because vech includes the diagonal. You
> can modify vech in order to discard even the diagonal equations.
>
> hth



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From tmacchant at yahoo.co.jp  Thu Sep 10 10:06:08 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Fri, 11 Sep 2009 00:06:08 +0900 (JST)
Subject: Netlab on Octave
In-Reply-To: <91e5dab20909100718g49d98773sfc70abb694b45aad@mail.gmail.com>
Message-ID: <20090910150608.28418.qmail@web3807.mail.bbt.yahoo.co.jp>

Hello

Please show us what is Netlab more in detail.

There wrote no information about demnlab.m.

What you think common sense is not always popular to other people.
You first reconsider your question again to explain the Netlab to those who did not have knowledge of
them.

BTW, why do you want use old octave ?
The compatibility to MATLAB is much improved in the octave 3.2 than that of octave 3.0.
 
Regards

Tatsuro   


--- Wilson de Oliveira  wrote:

> Hi. I hope anyone may help me
> As you may guess I am a complete newbie in Octave,
> but I need to run Netlab and have no access to Matlab.
> 
> I am using Gnu Octave for Windows 3.0.0.
> When trying to use the Netlab 3.3 demo ?demnlab?
> I get the following messages:
> 
> octave-3.0.0.exe:11> demnlab
> demnlab demnlab.m
> octave-3.0.0.exe:11> demnlab
> warning: set: invalid property `Name'
> warning: set: invalid property `NumberTitle'
> error: `uicontrol' undefined near line 33 column 3
> error: evaluating if command near line 20, column 1
> error: called from `demnlab' in file `c:\octave\Netlab\demnlab.m'
> octave-3.0.0.exe:11>
> 
> .
> Wilson de Oliveira
> UFRPE-DEInfo
> http://www.cin.ufpe.br/~wrdo
> > _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From dfslezak at gmail.com  Thu Sep 10 11:04:03 2009
From: dfslezak at gmail.com (=?ISO-8859-1?Q?Diego_Fern=E1ndez_Slezak?=)
Date: Thu, 10 Sep 2009 12:04:03 -0400
Subject: Compiling error
Message-ID: <cfd0f9eb0909100904i30690610rb971245b416c838b@mail.gmail.com>

Hello,

I'm trying to compile Octave in BlueGene.
My configure line is

./configure --disable-readline --disable-shared --disable-dl
--enable-static --target=ppc-bgp-gnu --host=ppc-bgp-gnu --without-hdf5
--without-fftw --without-x --with-x=no --without-readline
--without-gnuplot --with-blas=esslbg

When I try to compile it, it fails with the following error.
Apparently, the option --without-x and --with-x=no are not being taken
into account.
When I check the Makeconf file, I find HAVE_X_WINDOWS defined.
Any ideas?
.
.
.
/bgsys/drivers/ppcfloor/gnu-linux//bin/powerpc-bgp-linux-g++ -c
-I/bgsys/drivers/ppcfloor//include  -I. -I.. -I../liboctave -I../src
-I../libcruft/misc  -DHAVE_CONFIG_H  -Wall -W -Wshadow
-Wold-style-cast -Wformat -I/bgsys/drivers/ppcfloor//include -O3
display.cc -o display.o
display.cc:34:22: error: X11/Xlib.h: No such file or directory
display.cc: In member function ?void display_info::init()?:
display.cc:96: error: ?Display? was not declared in this scope
display.cc:96: error: ?display? was not declared in this scope
display.cc:96: error: ?XOpenDisplay? was not declared in this scope
display.cc:100: error: ?Screen? was not declared in this scope
display.cc:100: error: ?screen? was not declared in this scope
display.cc:100: error: ?DefaultScreenOfDisplay? was not declared in this scope
display.cc:104: error: ?DefaultDepthOfScreen? was not declared in this scope
display.cc:106: error: ?HeightOfScreen? was not declared in this scope
display.cc:107: error: ?WidthOfScreen? was not declared in this scope
display.cc:109: error: ?XScreenNumberOfScreen? was not declared in this scope
display.cc:111: error: ?DisplayHeightMM? was not declared in this scope
display.cc:112: error: ?DisplayWidthMM? was not declared in this scope
make[2]: *** [display.o] Error 1
make[2]: Leaving directory `/bgpusr2/xdslezak/octave-3.2.2/src'
make[1]: *** [src] Error 2
make[1]: Leaving directory `/bgpusr2/xdslezak/octave-3.2.2'
make: *** [all] Error 2


From udippel at uniten.edu.my  Thu Sep 10 11:39:21 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Fri, 11 Sep 2009 00:39:21 +0800
Subject: Help-octave Digest, Vol 42, Issue 18
In-Reply-To: <mailman.1241.1252594697.21430.help-octave@octave.org>
References: <mailman.1241.1252594697.21430.help-octave@octave.org>
Message-ID: <4AA92BB9.9030501@uniten.edu.my>

help-octave-request at octave.org wrote:
>
> Message: 2
> Date: Thu, 10 Sep 2009 01:19:17 -0400
> From: "John W. Eaton" <jwe at octave.org>
> Subject: Octave and Eidors
> To: Uwe Dippel <udippel at uniten.edu.my>
> Cc: help-octave at octave.org
> Message-ID: <19112.35925.323307.343873 at segfault.lan>
> Content-Type: text/plain; charset=us-ascii
>
> On 10-Sep-2009, Uwe Dippel wrote:
>
> | I am a total newbie. Not to Unix nor Linux, but to Octave and Eidors. I
> | followed what is written on the main site of eidors, but the examples fail:
> |
> | octave:1> run downloads/eidors-v3.3/startup.m
> | warning: isstr is obsolete and will be removed from a future version of
> | Octave, please use ischar instead
> | EIDORS:[Completed setting up of EIDORS Version 3.3 ($Date: 2008/07/19
> | 16:14:18 $)]
> | EIDORS:[Parameter: cache_size=100 MB]
> | EIDORS:[Parameter: mapped_colour=127]
> | EIDORS:[Default background colour: write]
> | EIDORS:[EIDORS mex folder:
> | /home/udippel/downloads/eidors-v3.3/arch/octave/pc]
> | octave:3> run downloads/eidors-v3.3/examples/demo_real.m
> | step 1: create FEM model structure
> | step 2: create FEM model electrodes definitions
> | step 3: create FEM model stimulation and measurement patterns
> | step 4: simulate data for homogeneous medium
> | step 5: simulate data for inhomogeneous medium
> | warning: set: invalid property `Name'
> | warning: set: invalid property `EdgeColor'
> | error: invalid value = Direct
> | error: called from `__patch__' in file
> | /usr/share/octave/3.0.1/m/plot/__patch__.m near line 182, column 5
>
> You might have better luck with graphics if you try the current Octave
> release, 3.2.2.
>   

Okay, that's the first step. How would this be done most favourably on 
Ubuntu 9.04? There is a download link that mentions packages 
(http://octave.sf.net/), but none is to be seen. Of course, I'd prefer a 
Ubuntu-.deb.

>
> Try using a newer version of Octave. Version 3.2.2 for example.. I
> suspect this will work there.. Though you should report to Eidors that
> "isstr" is deprecated in Octave
>   

They have changed it in their CVS already; so any next version should 
have it. I also noticed it in my failed efforts, but it was a warning only.

Uwe


From carlo.defalco at gmail.com  Thu Sep 10 11:57:48 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Thu, 10 Sep 2009 18:57:48 +0200
Subject: 4-D Array
Message-ID: <611AD04F-B050-4A54-8625-BF253463EC56@gmail.com>

Hi,
I am trying to create a 4D NDArray (3x3x3x3) from within
an oct-file but I don't understand how dim_vector works,
I tried the following but it crashes Octave:

   dim_vector a (4);
   a(0)=3;a(1)=3;a(2)=3;a(3)=3;
   NDArray b (a);

with the error:

../liboctave/dim-vector.h:113: failed assertion `i >= 0 && i < ndims'

any help appreciated,
c.


From jwe at octave.org  Thu Sep 10 12:09:21 2009
From: jwe at octave.org (John W. Eaton)
Date: Thu, 10 Sep 2009 13:09:21 -0400
Subject: 4-D Array
In-Reply-To: <611AD04F-B050-4A54-8625-BF253463EC56@gmail.com>
References: <611AD04F-B050-4A54-8625-BF253463EC56@gmail.com>
Message-ID: <19113.12993.833814.388335@segfault.lan>

On 10-Sep-2009, Carlo de Falco wrote:

| I am trying to create a 4D NDArray (3x3x3x3) from within
| an oct-file but I don't understand how dim_vector works,
| I tried the following but it crashes Octave:
| 
|    dim_vector a (4);

Looking at the dim-vector.h header file, I see

  explicit dim_vector (octave_idx_type n)
    : rep (newrep (2))
    {
      rep[0] = n;
      rep[1] = 1;
    }

so dim_vector a(4) creates a dimension vector for an 4x1 object.  You
need to do

  dim_vector a;
  a.resize (4);
  a(0) = 3;
  a(1) = 3;
  a(2) = 3;
  a(3) = 3;

or, since your dimensions are all the same, this should also work:

  dim_vector a;
  a.resize (4, 3);

jwe

From gkousiou at telecom.ntua.gr  Thu Sep 10 12:10:00 2009
From: gkousiou at telecom.ntua.gr (George Kousiouris)
Date: Thu, 10 Sep 2009 20:10:00 +0300 (EEST)
Subject: Unable to load matlab-produced mat files in octave
Message-ID: <1149.147.102.19.192.1252602600.squirrel@my.telecom.ntua.gr>


Hi all,

I have a neural network created with Matlab 7.5 and stored in a mat file.
I am trying to load it to Octave 3.0.1 unsuccessfully. I have also tried
saving it as older Matlab version, however with the older versions I get
the error

Unknown object.

Saving it as 7.3 version, I get the following error when attempting to
load it in Octave:

error:load:error while reading hdf5 item IW


Any suggestions?

Thanks in advance and best regards,
George





From Pascal.Dupuis at worldonline.be  Thu Sep 10 11:05:52 2009
From: Pascal.Dupuis at worldonline.be (Dupuis)
Date: Thu, 10 Sep 2009 09:05:52 -0700 (PDT)
Subject: sqp, objective and equality constraint gradients ?
Message-ID: <25385728.post@talk.nabble.com>


Hello,
I have troubles getting sqp to work. Problem is defined as finding points on
an ellipse verifying some optimum condition.
objective: f =@(x) -prod(x);
gradient:  fg = @(x) -[x(2); x(1)];
equality: ce = @(x) (x-center).'*M*(x-center) - level
gradient: ceg = @(x) 2*M*(x-center)
where x is a (2, 1) vector, center is a (2, 1) vector with the center
coordinates, M is symetric, positive definite and level the sum of square of
the distances between each point and the two ellipse defining points.
The call is as follows:
1) find some good candidate by  trying a few points, let's call it xinit
2) lb = .95*xinit; ub = 1.05*xinit %# define a five percent search interval
3)  [xopt, obj, info, iter, nf, lambda] = sqp(xinit, {f fg}, {ce ceg}, [],
lb, ub);

The problems are at line 343 and 377
  F = feval (ce_grd, x);
  C = feval (ci_grd, x);
  A = [F; C];
F is supposed to be (n, 1); while C is (2*n, n): they are not compatible ...
I tried transposing F or padding F with zeroes : if line 343 succeeds, the
code fails elsewhere, f.i. in line 377:  t4 = norm (lambda .* con) which
expects lambda (rows(F), 1) and con (= [ce; ci]) to be the same dimensions.
 Could someone please tell me the expected dimensions of the various
gradients ?

Regards

Pascal
-- 
View this message in context: http://www.nabble.com/sqp%2C-objective-and-equality-constraint-gradients---tp25385728p25385728.html
Sent from the Octave - General mailing list archive at Nabble.com.


From redays at gmail.com  Thu Sep 10 12:27:35 2009
From: redays at gmail.com (Leandro)
Date: Thu, 10 Sep 2009 14:27:35 -0300
Subject: Install 3.2.2
Message-ID: <bfb8167b0909101027q1124abd1occc4c704ed9fc92@mail.gmail.com>

When I do the 'make install' command It says that it can't find the file
conf.texi, but I type 'conf.texi.in' and it keeps going... but then It can't
find the file 'polynomial.m', but the file IS in the octave3.2.2 folder.
What do I have to do?

I am using Ubuntu 9.04 64 bits.

Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090910/d6f78626/attachment.html 

From ct529 at york.ac.uk  Thu Sep 10 12:53:16 2009
From: ct529 at york.ac.uk (Corrado)
Date: Thu, 10 Sep 2009 18:53:16 +0100
Subject: Install 3.2.2
In-Reply-To: <bfb8167b0909101027q1124abd1occc4c704ed9fc92@mail.gmail.com>
References: <bfb8167b0909101027q1124abd1occc4c704ed9fc92@mail.gmail.com>
Message-ID: <200909101853.16177.ct529@york.ac.uk>

It sounds strange .... did you type "sudo make install" or are you trying to 
install in the user home directory?

On Thursday 10 September 2009 18:27:35 Leandro wrote:
> When I do the 'make install' command It says that it can't find the file
> conf.texi, but I type 'conf.texi.in' and it keeps going... but then It
> can't find the file 'polynomial.m', but the file IS in the octave3.2.2
> folder. What do I have to do?
>
> I am using Ubuntu 9.04 64 bits.
>
> Thanks.



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From jwe at octave.org  Thu Sep 10 13:02:19 2009
From: jwe at octave.org (John W. Eaton)
Date: Thu, 10 Sep 2009 14:02:19 -0400
Subject: Install 3.2.2
In-Reply-To: <bfb8167b0909101027q1124abd1occc4c704ed9fc92@mail.gmail.com>
References: <bfb8167b0909101027q1124abd1occc4c704ed9fc92@mail.gmail.com>
Message-ID: <19113.16171.550556.867755@segfault.lan>

On 10-Sep-2009, Leandro wrote:

| When I do the 'make install' command It says that it can't find the file
| conf.texi, but I type 'conf.texi.in' and it keeps going... but then It can't
| find the file 'polynomial.m', but the file IS in the octave3.2.2 folder.
| What do I have to do?

Precisely what did you do?  Did you run configure?  Did you run make?
Or did you just unpack the tar file and run "make install"?

jwe

From dbateman at dbateman.org  Thu Sep 10 13:13:56 2009
From: dbateman at dbateman.org (David Bateman)
Date: Thu, 10 Sep 2009 20:13:56 +0200
Subject: Unable to load matlab-produced mat files in octave
In-Reply-To: <1149.147.102.19.192.1252602600.squirrel@my.telecom.ntua.gr>
References: <1149.147.102.19.192.1252602600.squirrel@my.telecom.ntua.gr>
Message-ID: <4AA941E4.4030001@dbateman.org>

George Kousiouris wrote:
> Hi all,
>
> I have a neural network created with Matlab 7.5 and stored in a mat file.
> I am trying to load it to Octave 3.0.1 unsuccessfully. I have also tried
> saving it as older Matlab version, however with the older versions I get
> the error
>
> Unknown object.
>
> Saving it as 7.3 version, I get the following error when attempting to
> load it in Octave:
>
> error:load:error while reading hdf5 item IW
>
>
> Any suggestions?
>
> Thanks in advance and best regards,
> George
>
>
>
>   

Matlab v7.3 format is not yet supported. Adding this support would make 
a nice project for someone, and was one of the things on my personal 
wishlist before I had to cut back my contributions to Octave. For now 
save your files in matlab with the the "-v7" flag or change your default 
file format preference as shown on the page

http://www.mathworks.com/access/helpdesk/help/techdoc/ref/save.html

D.

-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From carlo.defalco at gmail.com  Thu Sep 10 14:15:40 2009
From: carlo.defalco at gmail.com (Carlo de Falco)
Date: Thu, 10 Sep 2009 21:15:40 +0200
Subject: 4-D Array
In-Reply-To: <611AD04F-B050-4A54-8625-BF253463EC56@gmail.com>
References: <611AD04F-B050-4A54-8625-BF253463EC56@gmail.com>
Message-ID: <A3C7C266-74B0-4B18-A7FD-E8136C95FF4C@gmail.com>


On 10 Sep 2009, at 18:57, Carlo de Falco wrote:

> Hi,
> I am trying to create a 4D NDArray (3x3x3x3) from within
> an oct-file but I don't understand how dim_vector works,
> I tried the following but it crashes Octave:
>
>  dim_vector a (4);
>  a(0)=3;a(1)=3;a(2)=3;a(3)=3;
>  NDArray b (a);
>
> with the error:
>
> ../liboctave/dim-vector.h:113: failed assertion `i >= 0 && i < ndims'
>
> any help appreciated,
> c.
>

ooops looking at the constructor implementation I see that
"dim_vector a (4)" is equivalent to "dim_vector a (4, 1)",
so the above makes no sense...
c.


From highegg at gmail.com  Thu Sep 10 14:36:07 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Thu, 10 Sep 2009 21:36:07 +0200
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <200909101557.47893.ct529@york.ac.uk>
References: <200909071338.28862.ct529@york.ac.uk>
	<200909071521.39612.ct529@york.ac.uk>
	<69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
	<200909101557.47893.ct529@york.ac.uk>
Message-ID: <69d8d540909101236j4f00c45cl622b06f0d9e727e1@mail.gmail.com>

On Thu, Sep 10, 2009 at 4:57 PM, Corrado <ct529 at york.ac.uk> wrote:
> Dear Jaroslav, Octave list,
>
> here I am again ....
>
> How do you calculate distance or dissimilarity matrices in Octave?

For instance, see
http://octave.sourceforge.net/doc/f/pdist.html

> and how do
> you treat triangular matrices + diagonal in Octave? ?is there a separate
> package?
>

That depends. Diagonal matrices are a special type.
Triangular/symmetric matrices are typically treated as full, with
certain functions (such as matrix division) able to specialize. Of
course, you can also work with vectors of the upper/lower triangular
elements, or even with "jagged arrays" - cell arrays of vectors. Your
choice.

There's also a package for triangular matrices in OctaveForge's
extra/, but I don't think it's much used.

hth
-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From dbateman at dbateman.org  Thu Sep 10 15:06:33 2009
From: dbateman at dbateman.org (David Bateman)
Date: Thu, 10 Sep 2009 22:06:33 +0200
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <69d8d540909101236j4f00c45cl622b06f0d9e727e1@mail.gmail.com>
References: <200909071338.28862.ct529@york.ac.uk>	<200909071521.39612.ct529@york.ac.uk>	<69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>	<200909101557.47893.ct529@york.ac.uk>
	<69d8d540909101236j4f00c45cl622b06f0d9e727e1@mail.gmail.com>
Message-ID: <4AA95C49.1060604@dbateman.org>

Jaroslav Hajek wrote:
> On Thu, Sep 10, 2009 at 4:57 PM, Corrado <ct529 at york.ac.uk> wrote:
>   
>> Dear Jaroslav, Octave list,
>>
>> here I am again ....
>>
>> How do you calculate distance or dissimilarity matrices in Octave?
>>     
>
> For instance, see
> http://octave.sourceforge.net/doc/f/pdist.html
>
>   
>> and how do
>> you treat triangular matrices + diagonal in Octave?  is there a separate
>> package?
>>
>>     
>
> That depends. Diagonal matrices are a special type.
> Triangular/symmetric matrices are typically treated as full, with
> certain functions (such as matrix division) able to specialize. Of
> course, you can also work with vectors of the upper/lower triangular
> elements, or even with "jagged arrays" - cell arrays of vectors. Your
> choice.
>
> There's also a package for triangular matrices in OctaveForge's
> extra/, but I don't think it's much used.
>
> hth
>   
The triangular matrix code is in octave-forge for the sole reason of 
having an example user defined type in C++. See the comments in the docs

http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/extra/triangular/doc/triangular.texi?revision=HEAD&view=markup

So not surprising the triangular class in octave-forge isn't "used", as 
its use is as an example for others.

D.


From vic at norton.name  Thu Sep 10 16:29:30 2009
From: vic at norton.name (Vic Norton)
Date: Thu, 10 Sep 2009 17:29:30 -0400
Subject: Mac OSX 10.6 and Octave
Message-ID: <202B0033-5864-4FC4-AD8A-973756E1FFB7@norton.name>

Has anyone been able to successfully install Octave 3.2.2 (or any  
earlier version) on the new Mac OSX 10.6 (Snow Leopard) operating  
system? I've heard nothing but horror stories so far. I would very  
much like to upgrade to Mac OSX 10.6 (which I have purchased), but I  
dare not upgrade unless I can install Octave. Mind you, I am an old  
(74), unix-un-savvy mathematician; I need a relatively straightforward  
installation process.

By the way, Octave 3.2.2 works perfectly now, on my 2.93 GHz Intel  
Core 2 Duo iMac with Mac OSX 10.5.8. As I understand it, Octave 3.2.2  
will stop working as soon as I upgrade to Mac OSX 10.6. Am I wrong  
about this?

Any advice would be greatly appreciated.

Regards,

Vic


From bpabbott at mac.com  Thu Sep 10 17:09:48 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Thu, 10 Sep 2009 18:09:48 -0400
Subject: Mac OSX 10.6 and Octave
In-Reply-To: <202B0033-5864-4FC4-AD8A-973756E1FFB7@norton.name>
References: <202B0033-5864-4FC4-AD8A-973756E1FFB7@norton.name>
Message-ID: <C389F8CA-0C9D-44E7-9620-7D39D357B933@mac.com>

On Sep 10, 2009, at 5:29 PM, Vic Norton wrote:

> Has anyone been able to successfully install Octave 3.2.2 (or any
> earlier version) on the new Mac OSX 10.6 (Snow Leopard) operating
> system? I've heard nothing but horror stories so far. I would very
> much like to upgrade to Mac OSX 10.6 (which I have purchased), but I
> dare not upgrade unless I can install Octave. Mind you, I am an old
> (74), unix-un-savvy mathematician; I need a relatively straightforward
> installation process.
>
> By the way, Octave 3.2.2 works perfectly now, on my 2.93 GHz Intel
> Core 2 Duo iMac with Mac OSX 10.5.8. As I understand it, Octave 3.2.2
> will stop working as soon as I upgrade to Mac OSX 10.6. Am I wrong
> about this?
>
> Any advice would be greatly appreciated.
>
> Regards,
>
> Vic

I'm presently unable to build octave, but have no problem running the  
binary at the link below.

	http://sourceforge.net/projects/octave/files/Octave%20MacOSX%20Binary

However, if you need to compile oct-files, you'll want to wait until  
the problem with 'mkoctfile' is fixed.

	http://www.nabble.com/Install-packages-in-Mac-OS-X-td25356552.html

Ben


From tmacchant at yahoo.co.jp  Thu Sep 10 19:19:04 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Fri, 11 Sep 2009 09:19:04 +0900 (JST)
Subject: Netlab on Octave
In-Reply-To: <91e5dab20909100829r19723a32teb8d9f19c151b0f3@mail.gmail.com>
Message-ID: <20090911001904.22946.qmail@web3802.mail.bbt.yahoo.co.jp>

Hello Wilson de Oliveira

--- Wilson de Oliveira wrote:

> Ok. Sorry.
>  For a start, Netlab is a Matlab toolbox for Pattern Recognition which
> includes amongst others models Neural Networks. It is freely available at:
> http://www.ncrg.aston.ac.uk/netlab/index.php
> demnlab.m is a demonstration program which runs several models in gui
> interface.
> 
> As regard the version of the octave, thanks for your help! The octave 3.0.0
> came with qtoctave - a gui interface for octave; Now that I?ve installed the
> 3.2.2 version available at sourceforge.net I got fewer errors:
> 
>  > demnlab
> error: `uicontrol' undefined near line 33 column 3
> error: called from:
> error:   C:\Octave\Netlab\demnlab.m at line 33, column 3
> octave-3.2.2.exe:6:C:\Octave\Netlab
> >
> 
> Thanks for your help
> Wilson de Oliveira
> UFRPE-DEInfo
> http://www.cin.ufpe.br/~wrdo
> 
> 

Unfortunately Octave does not support GUI graphic control.
You should control graphics with graphic handle functions.
Please  see section 15 "Plotting" in octave manual.
You should read care fully "15.1 Plotting Basics" band "15.2 Advanced Plotting".
Perhaps these types functions are compatible to MATLAB.


I have tried a simple example in web page.

ndata=20
% Generate the matrix of inputs x and targets t.
x = [0:1/19:1]';
t = sin(2*pi*x) + 0.2*randn(ndata, 1);

% Set up network parameters.
net = mlp(1, 3, 1, 'linear')

net =
{
  type = mlp
  nin =  1
  nhidden =  3
  nout =  1
  nwts =  10
  outfn = linear
  w1 =

    -1.148633   1.104356   0.049478

  b1 =

    -0.070054   0.902782   1.349561

  w2 =

     0.17895
     0.27658
     0.36492

  b2 =  0.66142
}
****************
% Set up vector of options for the optimiser.
options = zeros(1,18);
options(1) = 1;
options(9) = 1;
options(14) = 100;

% Train using scaled conjugate gradients.
[net, options] = netopt(net, options, x, t, 'scg');
error: `fcnchk' undefined near line 68 column 5
error: called from:
error:   D:\usr\Tatsu\program\Octave\netlab\scg.m at line 68, column 3
error:   D:\usr\Tatsu\program\Octave\netlab\netopt.m at line 36, column 17

Octave does not support `fcnchk'
But I found 
http://en.wikibooks.org/wiki/MATLAB_Programming/Differences_between_Octave_and_MATLAB

********
If something (like Netlab) need a function called fcnchk you can just put the following into a file
called fcnchk.m and put it somewhere Octave can find it:
function f=fcnchk(x, n)
  f = x;
end
********
I placed fcnchk.m under the same folder of netlab

SO I get 

Checking gradient ...

   analytic   diffs     delta

  1.1904e+000  1.1904e+000  -1.3890e-009
  4.8963e-001  4.8963e-001  8.2556e-010
  1.1238e+000  1.1238e+000  -5.8918e-010
  2.3292e+000  2.3292e+000  -1.3039e-010
  1.1362e+000  1.1362e+000  -1.2848e-009
  1.8523e+000  1.8523e+000  6.3412e-010
  -1.3883e+001  -1.3883e+001  -1.9210e-010
  2.0438e+001  2.0438e+001  5.2320e-010
  1.9960e+001  1.9960e+001  -9.0928e-011
  2.2659e+001  2.2659e+001  -8.4377e-010
Cycle    1  Error    4.629483  Scale 1.000000e+000
Cycle    2  Error    4.629483  Scale 5.000000e-001
Cycle    3  Error    4.629483  Scale 2.000000e+000
Cycle    4  Error    4.259136  Scale 8.000000e+000
Cycle    5  Error    4.050134  Scale 4.000000e+000
Cycle    6  Error    3.775366  Scale 2.000000e+000
Cycle    7  Error    3.390233  Scale 2.000000e+000
Cycle    8  Error    3.288238  Scale 1.000000e+000
Cycle    9  Error    3.286530  Scale 5.000000e-001
Cycle   10  Error    3.235898  Scale 2.500000e-001
Cycle   11  Error    3.212532  Scale 1.250000e-001
Cycle   12  Error    3.192542  Scale 6.250000e-002
Cycle   13  Error    3.187294  Scale 3.125000e-002
  :
  :
*****************************************
% Plot the trained network predictions.
plotvals = [0:0.01:1]';
y = mlpfwd(net, plotvals);
plot(plotvals, y, 'ob')

gives a plot like attatchment files.

********
Honestly I am not familiar with this type of work.
In addition, I have no experience of MATLAB

Hopefully users who use both MATLAB and octave give you much better suggestions.

The aboves are all what I can do for you.

Good luck!!

Regards

Tatsuro
 

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sample.png
Type: image/png
Size: 8835 bytes
Desc: 3472330463-sample.png
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/d3ebef1b/attachment.png 

From Potorti at isti.cnr.it  Fri Sep 11 02:21:47 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Fri, 11 Sep 2009 09:21:47 +0200
Subject: Matrix regression of distnce matrices + non negative least square
In-Reply-To: <200909101557.47893.ct529@york.ac.uk>
References: <200909071338.28862.ct529@york.ac.uk>
	<200909071521.39612.ct529@york.ac.uk>
	<69d8d540909070804i7f1d5821v9c5a4e758dd2338a@mail.gmail.com>
	<200909101557.47893.ct529@york.ac.uk>
Message-ID: <E1Mm0S3-0003IL-9a@tucano.isti.cnr.it>

>How do you calculate distance or dissimilarity matrices in Octave?

Maybe the statistics package can help you?  Have you looked at pdist?

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From highegg at gmail.com  Fri Sep 11 04:17:16 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 11:17:16 +0200
Subject: new function for quadratic programming: pqpnonneg
Message-ID: <69d8d540909110217j423527ebrf6fe14f2b6b321a9@mail.gmail.com>

hi all,

I just contributed a new function: pqpnonneg (Positive Quadratic
Programming in NONNEGative variables).
It can be used to solve problems of the form

min 1/2*x'*A*x + b'*x, all (x >= 0)
where A is a positive definite matrix. The implementation exploits a
duality between least squares and positive qp problems; it is very
similar to lsqnonneg except that it solves the dual system A \ -b and
works with a Cholesky factorization instead of QR (via qrinsert &
qrdelete).

If applicable, there are two advantages against qp:

1. it is usually significantly faster
2. it will always converge in a finite number of iterations (I think
this follows from the Lawson-Hanson proof and the duality)

both of these are of course very desirable. Here's a small benchmark
creating an artificial problem and solving it via both pqpnonneg and
qp:

n = 500;
## Generate SPD matrix
for i = 1:3
  A = rand (n);
  A = A'*A + 1e-3*eye (n);
  x = rand (n, 1) - 0.5;
  b = -(A*x);
  disp ("pqpnonneg")
  tic;
  [x1, obj1] = pqpnonneg (A, b);
  toc
  norm (x1 - x)

  disp ("qp")
  tic;
  [x1, obj1] = qp (zeros (n, 1), A, b, [], [], zeros (n, 1), []);
  toc
  norm (x1 - x)
  disp ("--------------------------")
endfor

On my Core 2 Duo @ 2.83 GHz, g++ -O3 -march=native + serial GotoBLAS, I got:

pqpnonneg
Elapsed time is 0.00449514 seconds.
ans =  6.2362
qp
Elapsed time is 1.162 seconds.
ans =  6.2362
--------------------------
pqpnonneg
Elapsed time is 0.001014 seconds.
ans =  6.7428
qp
Elapsed time is 1.144 seconds.
ans =  6.7428
--------------------------
pqpnonneg
Elapsed time is 0.02081 seconds.
ans =  6.2335
qp
Elapsed time is 32.09 seconds.
ans =  6.2335
--------------------------
pqpnonneg
Elapsed time is 0.0235 seconds.
ans =  6.0290
qp
Elapsed time is 35.54 seconds.
ans =  6.0290
--------------------------
pqpnonneg
Elapsed time is 0.02568 seconds.
ans =  6.1129
qp
Elapsed time is 37.88 seconds.
ans =  6.1129
--------------------------

so, as you can see, pqpnonneg can be many times faster, as much as
1500x in this benchmark.
Btw. the reason why this goes into Octave rather than
OctaveForge/optim is the duality with lsqnonneg. If ever lsqnonneg is
moved to optim, so should pqpnonneg.

enjoy!

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Fri Sep 11 05:07:25 2009
From: ct529 at york.ac.uk (Corrado)
Date: Fri, 11 Sep 2009 11:07:25 +0100
Subject: CUDA  in Octave 322
Message-ID: <200909111107.25284.ct529@york.ac.uk>

Dear Octave list,

is there any support for CUDA  in Octave 322?

Best,
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From ct529 at york.ac.uk  Fri Sep 11 05:11:56 2009
From: ct529 at york.ac.uk (Corrado)
Date: Fri, 11 Sep 2009 11:11:56 +0100
Subject: Benchmarking Octave 322
Message-ID: <200909111111.57102.ct529@york.ac.uk>

Dear list,

I have compiled 322 with and without option -march=native on gcc 4.3.3 on 
kubuntu64 bit, and I would like to test which is more performant. Is there a 
standard Octave benchmark I could use?

Regards
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Fri Sep 11 05:17:04 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 12:17:04 +0200
Subject: CUDA in Octave 322
In-Reply-To: <200909111107.25284.ct529@york.ac.uk>
References: <200909111107.25284.ct529@york.ac.uk>
Message-ID: <69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>

On Fri, Sep 11, 2009 at 12:07 PM, Corrado <ct529 at york.ac.uk> wrote:
> Dear Octave list,
>
> is there any support for CUDA ?in Octave 322?
>
> Best,
> --
> Corrado Topi
>
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>

In short, no. There are codes for Matlab that *might* work with Octave
(possibly after modifications). You can, of course, use CUDA in your
own C++ extensions, but you can't distribute those, because CUDA is
not free software.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From highegg at gmail.com  Fri Sep 11 05:21:04 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 12:21:04 +0200
Subject: Benchmarking Octave 322
In-Reply-To: <200909111111.57102.ct529@york.ac.uk>
References: <200909111111.57102.ct529@york.ac.uk>
Message-ID: <69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>

On Fri, Sep 11, 2009 at 12:11 PM, Corrado <ct529 at york.ac.uk> wrote:
> Dear list,
>
> I have compiled 322 with and without option -march=native on gcc 4.3.3 on
> kubuntu64 bit, and I would like to test which is more performant. Is there a
> standard Octave benchmark I could use?
>
> Regards
> --
> Corrado Topi
>
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>

There is the benchmark package at
http://octave.sourceforge.net/benchmark/index.html

and also some codes all over the web (mostly sadly out-of-date).

You can also invent your own benchmarks; in particular, contributions
to the benchmark package will be warmly welcome and coldly scrutinized
:).

regards

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Fri Sep 11 05:32:49 2009
From: ct529 at york.ac.uk (Corrado)
Date: Fri, 11 Sep 2009 11:32:49 +0100
Subject: CUDA in Octave 322
In-Reply-To: <69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
Message-ID: <200909111132.50005.ct529@york.ac.uk>

Sorry I had hoped nvidia would have released the driver FS/OS .... :(

I definitively do not want to use non FS/OS software .... do you know about the 
status of OpenCL over Nvidia or ATI?


On Friday 11 September 2009 11:17:04 Jaroslav Hajek wrote:
> On Fri, Sep 11, 2009 at 12:07 PM, Corrado <ct529 at york.ac.uk> wrote:
> > Dear Octave list,
> >
> > is there any support for CUDA  in Octave 322?
> >
> > Best,
> > --
> > Corrado Topi
> >
> > Global Climate Change & Biodiversity Indicators
> > Area 18,Department of Biology
> > University of York, York, YO10 5YW, UK
> > Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>
> In short, no. There are codes for Matlab that *might* work with Octave
> (possibly after modifications). You can, of course, use CUDA in your
> own C++ extensions, but you can't distribute those, because CUDA is
> not free software.



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Fri Sep 11 05:37:44 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 12:37:44 +0200
Subject: CUDA in Octave 322
In-Reply-To: <200909111132.50005.ct529@york.ac.uk>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
	<200909111132.50005.ct529@york.ac.uk>
Message-ID: <69d8d540909110337x7d433f4enb10ab03c63ad4f95@mail.gmail.com>

On Fri, Sep 11, 2009 at 12:32 PM, Corrado <ct529 at york.ac.uk> wrote:
> Sorry I had hoped nvidia would have released the driver FS/OS .... :(
>
> I definitively do not want to use non FS/OS software .... do you know about the
> status of OpenCL over Nvidia or ATI?
>

Personally, I'm fairly ignorant about GPGPU, primarily because I have
no GPU hardware around. You'll have to investigate yourself.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From Potorti at isti.cnr.it  Fri Sep 11 06:48:27 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Fri, 11 Sep 2009 13:48:27 +0200
Subject: Benchmarking Octave 322
In-Reply-To: <69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>
References: <200909111111.57102.ct529@york.ac.uk>
	<69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>
Message-ID: <E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>

>There is the benchmark package at
>http://octave.sourceforge.net/benchmark/index.html
>
>and also some codes all over the web (mostly sadly out-of-date).

In the past I used to maintain a benchmark.m, which is still available
at <http://fly.isti.cnr.it/pub/software/octave/benchmark.m>.  It still
works on Octave 3.2.2 and claims to compute the performance relative to
Octave 2 on a Pentium II at 350 MHz with Debian's lapack3 and refblas3
as of November 2005.  Its main asset is that it computes a 95%
confidence interval for the results, something that as far as I can see
is missing from your code.

I even used to publish results sent from people on
<http://fly.isti.cnr.it/pub/software/octave/bm_results>, back when the
reference was Sun Sparc, about 100 times slower than the current Pentium
II reference.  I stopped updating the results when compiling Octave
became so much dependent on libraries and compiling options that
producing a single number for a given hardware would have been
misleading, and maintaining a significant list would have been too
complex.

I often thought about working to merge it with your code, but never
found the time, and I find myself adding things in front of this on my
wishlist.  So I am writing this mail, in the hope that you, or someone
else, possiby with my help, can work towards a better and easy-to-use
benchmark.  It would also be nice to have a page on the Octave wiki with
some performance results.

For a reference, here is what I see using my benchmark on an AMD Athlon
64 X2 3800+ running Debian's Octave 3.2.2:

octave> benchmark
Octave benchmark version bm 2.3
Speed of octave 3.2.2 on x86_64-pc-linux-gnu
  Pentium II 350 MHz has score 1.0, higher numbers are better
Matrix inversion (LAPACK)         9.05 +/- 0.3% (154 runs)
Schur decomposition (LAPACK)     23.88 +/- 0.2% (126 runs)
Differential equation (LSODE)    22.83 +/- 0.3% (546 runs)
Fourier transforms (FFTPACK)      7.43 +/- 0.5% (196 runs)
for loop                         34.76 +/- 1.0% (658 runs)
-- Performance index (bm 2.3): 17

It is surprising to me to see that such a modern processor is less than
ten times faster on some tests than a Pentium II!

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From ilnseher at eit.uni-kl.de  Fri Sep 11 07:15:04 2009
From: ilnseher at eit.uni-kl.de (Thomas Ilnseher)
Date: Fri, 11 Sep 2009 14:15:04 +0200
Subject: CUDA in Octave 322
In-Reply-To: <69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
Message-ID: <1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>

Am Freitag, den 11.09.2009, 12:17 +0200 schrieb Jaroslav Hajek:
> You can, of course, use CUDA in your
> own C++ extensions, but you can't distribute those, because CUDA is
> not free software.
Ok, I might be wrong here, but AFAIK Octave uses LAPACK.

As GPL still allows linking against CS "platform libraries", you can
still link Octave against a CS CUDA enabled LAPACK Lib (granted that
such a thing exists) when you define LAPACK as a platform library. 

Same thing as running *ANY* GPL program on Windowze... (it will link
against some CS win32 libs dynamically).

Same thing as running a GPL prog on Linux w/ NV or AMD binary driver
installed.
... And So On (tm).

It's the question whether LAPACK is a platform library. IMHO it is (you
can substitute ATLAS with IMKL AMD something which are CSd).


Now you only need a LAPACK lib that exploits CUDA.

Problem here is that you have to  copy data to the GPU's memory nad
back, call the GPU and so on. This probably gives a reasonable overhead,
so you'd only accelerate things this way when operating with *LARGE*
vectors / matrices.
-- 
Thomas Ilnseher <ilnseher at eit.uni-kl.de>


From sedwards2 at cinci.rr.com  Fri Sep 11 07:31:14 2009
From: sedwards2 at cinci.rr.com (Stuart Edwards)
Date: Fri, 11 Sep 2009 08:31:14 -0400
Subject: Mac OSX 10.6 and Octave
In-Reply-To: <202B0033-5864-4FC4-AD8A-973756E1FFB7@norton.name>
References: <202B0033-5864-4FC4-AD8A-973756E1FFB7@norton.name>
Message-ID: <C53F02A1-55DB-4B6F-9BEC-B23CB212F720@cinci.rr.com>

I was running 3.2.0 when I upgraded to 10.6 and everything still seems  
to be working fine.

Stu
On Sep 10, 2009, at 5:29 PM, Vic Norton wrote:

> Has anyone been able to successfully install Octave 3.2.2 (or any
> earlier version) on the new Mac OSX 10.6 (Snow Leopard) operating
> system? I've heard nothing but horror stories so far. I would very
> much like to upgrade to Mac OSX 10.6 (which I have purchased), but I
> dare not upgrade unless I can install Octave. Mind you, I am an old
> (74), unix-un-savvy mathematician; I need a relatively straightforward
> installation process.
>
> By the way, Octave 3.2.2 works perfectly now, on my 2.93 GHz Intel
> Core 2 Duo iMac with Mac OSX 10.5.8. As I understand it, Octave 3.2.2
> will stop working as soon as I upgrade to Mac OSX 10.6. Am I wrong
> about this?
>
> Any advice would be greatly appreciated.
>
> Regards,
>
> Vic
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave


From highegg at gmail.com  Fri Sep 11 07:38:28 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 14:38:28 +0200
Subject: Benchmarking Octave 322
In-Reply-To: <E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>
References: <200909111111.57102.ct529@york.ac.uk>
	<69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>
	<E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>
Message-ID: <69d8d540909110538u296bc919neede8a1f9252f927@mail.gmail.com>

On Fri, Sep 11, 2009 at 1:48 PM, Francesco Potorti` <Potorti at isti.cnr.it> wrote:
>>There is the benchmark package at
>>http://octave.sourceforge.net/benchmark/index.html
>>
>>and also some codes all over the web (mostly sadly out-of-date).
>
> In the past I used to maintain a benchmark.m, which is still available
> at <http://fly.isti.cnr.it/pub/software/octave/benchmark.m>. ?It still
> works on Octave 3.2.2 and claims to compute the performance relative to
> Octave 2 on a Pentium II at 350 MHz with Debian's lapack3 and refblas3
> as of November 2005. ?Its main asset is that it computes a 95%
> confidence interval for the results, something that as far as I can see
> is missing from your code.
>

You're right, that could be of some use. I just added that capability.

> I even used to publish results sent from people on
> <http://fly.isti.cnr.it/pub/software/octave/bm_results>, back when the
> reference was Sun Sparc, about 100 times slower than the current Pentium
> II reference. ?I stopped updating the results when compiling Octave
> became so much dependent on libraries and compiling options that
> producing a single number for a given hardware would have been
> misleading, and maintaining a significant list would have been too
> complex.

The mistake is to present it as benchmarking the hardware. It's
benchmarking particular installations. Still, comparisons can be very
useful. A typical usage is to compare a generic distro package with a
custom compilation.

> I often thought about working to merge it with your code, but never
> found the time, and I find myself adding things in front of this on my
> wishlist. ?So I am writing this mail, in the hope that you, or someone
> else, possiby with my help, can work towards a better and easy-to-use
> benchmark. ?It would also be nice to have a page on the Octave wiki with
> some performance results.

The package is also low priority for me; besides, it seems almost
nobody uses it. The point was to provide a package compatible with
older Octaves, even Matlab, so that users can benchmark and share
results, and write new benchmarks.

Writing a new benchmark is easy:

% function benchmark_pinv (m,n)
% description:
% Benchmarking the pseudoinverse
%
% arguments:
% m = number of rows
% n = number of columns
%
% results:
% time_rand = time for pinv of a random m*n matrix
% time_diag = time for pinv of a random diagonal m*n matrix
%

function results = benchmark_pinv (m,n)

  benchutil_default_arg ('m', 400);
  benchutil_default_arg ('n', 400);

  benchutil_initialize (mfilename)

  a = rand (m, n);
  tic; pinv (a); time_rand = toc;
  benchutil_set_result ('time_rand')

  % Do it in a compatible way.
  a = diag (rand (max (m, n), 1));
  a = a(1:m, 1:n);
  tic; pinv (a); time_diag = toc;
  benchutil_set_result ('time_diag')


now, just run it in various ways:

octave:1> benchmark_pinv
Running benchmark: benchmark_pinv
Benchmarking the pseudoinverse

number of rows (m) = 400
number of columns (n) = 400
time for pinv of a random m*n matrix: 0.561242
time for pinv of a random diagonal m*n matrix: 2.00272e-05
octave:2> benchmark_pinv (600, 300)
Running benchmark: benchmark_pinv
Benchmarking the pseudoinverse

number of rows (m) = 600
number of columns (n) = 300
time for pinv of a random m*n matrix: 0.243105
time for pinv of a random diagonal m*n matrix: 1.40667e-05
octave:3> benchutil_average("benchmark_pinv", 10, 600, 300)
Running benchmark: benchmark_pinv
Benchmarking the pseudoinverse

number of rows (m) = 600
number of columns (n) = 300
time for pinv of a random m*n matrix: 0.237562
time for pinv of a random diagonal m*n matrix: 1.40667e-05

....

time for pinv of a random m*n matrix (avg. over 10 runs): 0.240214 +/- 1.595569%
time for pinv of a random diagonal m*n matrix (avg. over 10 runs):
0.000014 +/- 3.743879%

The descriptions are auto-extracted from the help string and progress
is displayed throughout each run, as well as auto-averaging over
multiple runs via benchutil_average.


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From ct529 at york.ac.uk  Fri Sep 11 07:53:32 2009
From: ct529 at york.ac.uk (Corrado)
Date: Fri, 11 Sep 2009 13:53:32 +0100
Subject: CUDA in Octave 322
In-Reply-To: <1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
	<1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
Message-ID: <200909111353.32334.ct529@york.ac.uk>

On Friday 11 September 2009 13:15:04 Thomas Ilnseher wrote:
> Problem here is that you have to  copy data to the GPU's memory nad
> back, call the GPU and so on. This probably gives a reasonable overhead,
> so you'd only accelerate things this way when operating with *LARGE*
> vectors / matrices.

Yes, this is the situation here (very large matrices). The fact that CUDA is 
not FS/OS is still a problem though. That was the reason for mentioning 
OpenCL.

Regards
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Fri Sep 11 07:54:47 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 11 Sep 2009 14:54:47 +0200
Subject: CUDA in Octave 322
In-Reply-To: <1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
	<1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
Message-ID: <69d8d540909110554y149f460bs1c2d14425f78aabc@mail.gmail.com>

On Fri, Sep 11, 2009 at 2:15 PM, Thomas Ilnseher <ilnseher at eit.uni-kl.de> wrote:
> Am Freitag, den 11.09.2009, 12:17 +0200 schrieb Jaroslav Hajek:
>> You can, of course, use CUDA in your
>> own C++ extensions, but you can't distribute those, because CUDA is
>> not free software.
> Ok, I might be wrong here, but AFAIK Octave uses LAPACK.
>
> As GPL still allows linking against CS "platform libraries", you can
> still link Octave against a CS CUDA enabled LAPACK Lib (granted that
> such a thing exists) when you define LAPACK as a platform library.
>
> Same thing as running *ANY* GPL program on Windowze... (it will link
> against some CS win32 libs dynamically).
>
> Same thing as running a GPL prog on Linux w/ NV or AMD binary driver
> installed.
> ... And So On (tm).
>
> It's the question whether LAPACK is a platform library. IMHO it is (you
> can substitute ATLAS with IMKL AMD something which are CSd).
>
>
> Now you only need a LAPACK lib that exploits CUDA.
>
> Problem here is that you have to ?copy data to the GPU's memory nad
> back, call the GPU and so on. This probably gives a reasonable overhead,
> so you'd only accelerate things this way when operating with *LARGE*
> vectors / matrices.
> --
> Thomas Ilnseher <ilnseher at eit.uni-kl.de>
>
>

It seems to me that you are confusing things.
You can link Octave to whatever you want to, including arbitrary
proprietary libraries. What you typically can't do is to distribute
the resulting binaries, if they are derivative works of the non-free
library used. Proprietary BLAS/LAPACK implementations certainly do not
qualify for the system library exception, but as long as you use
dynamic linking and only standard BLAS/LAPACK calls, there is no
problem because the binary will work with any BLAS/LAPACK.


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From romeyke at cbs.mpg.de  Fri Sep 11 08:05:18 2009
From: romeyke at cbs.mpg.de (Andreas Romeyke)
Date: Fri, 11 Sep 2009 15:05:18 +0200
Subject: CUDA in Octave 322
In-Reply-To: <1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
References: <200909111107.25284.ct529@york.ac.uk>
	<69d8d540909110317wa6cf48g8e2ed58cf9411ce0@mail.gmail.com>
	<1252671305.13983.24.camel@pcmik05.zmk.uni-kl.de>
Message-ID: <20090911150518.4bec1b75@linde.cbs.mpg.de>

Hello,

Am Fri, 11 Sep 2009 14:15:04 +0200
schrieb Thomas Ilnseher <ilnseher at eit.uni-kl.de>:

> Problem here is that you have to  copy data to the GPU's memory nad
> back, call the GPU and so on. This probably gives a reasonable
> overhead, so you'd only accelerate things this way when operating
> with *LARGE* vectors / matrices.

Nope, you would only accelerate things this way when *heavily*
operating on vectors/matrices. The bottleneck is the data transfer
between GPU and CPU. There are limits in branching, there are limits in
jumping or random access the video memory.

Bye andreas

-- 
Software Developer / Dipl. Inform. (FH)
Max Planck Institute for Human Cognitive and Brain Sciences
Department of Psychology
Stephanstr. 1a, 04103 Leipzig, Germany

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/61528433/attachment.bin 

From dasergatskov at gmail.com  Fri Sep 11 08:48:14 2009
From: dasergatskov at gmail.com (Dmitri A. Sergatskov)
Date: Fri, 11 Sep 2009 08:48:14 -0500
Subject: Benchmarking Octave 322
In-Reply-To: <E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>
References: <200909111111.57102.ct529@york.ac.uk>
	<69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>
	<E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>
Message-ID: <892b16670909110648w63864372mece42ba6769c2363@mail.gmail.com>

On Fri, Sep 11, 2009 at 6:48 AM, Francesco Potorti` <Potorti at isti.cnr.it> wrote:

> For a reference, here is what I see using my benchmark on an AMD Athlon
> 64 X2 3800+ running Debian's Octave 3.2.2:
>
> octave> benchmark
> Octave benchmark version bm 2.3
> Speed of octave 3.2.2 on x86_64-pc-linux-gnu
> ?Pentium II 350 MHz has score 1.0, higher numbers are better
> Matrix inversion (LAPACK) ? ? ? ? 9.05 +/- 0.3% (154 runs)
> Schur decomposition (LAPACK) ? ? 23.88 +/- 0.2% (126 runs)
> Differential equation (LSODE) ? ?22.83 +/- 0.3% (546 runs)
> Fourier transforms (FFTPACK) ? ? ?7.43 +/- 0.5% (196 runs)
> for loop ? ? ? ? ? ? ? ? ? ? ? ? 34.76 +/- 1.0% (658 runs)
> -- Performance index (bm 2.3): 17
>

It looks to me your ATLAS is not well optimized. Here is 3.2.2 on
Fedora 11 x86_64
(Opteron 180, at 2.4GHz):

Matrix inversion (LAPACK)        31.47 +/- 2.6% (539 runs)
Schur decomposition (LAPACK)     39.68 +/- 0.1% (238 runs)
Differential equation (LSODE)    26.19 +/- 0.6% (623 runs)
Fourier transforms (FFTPACK)      7.09 +/- 1.9% (168 runs)
for loop                         35.66 +/- 1.1% (665 runs)
-- Performance index (bm 2.3): 24


> --
> Francesco Potort? (ricercatore) ? ? ? ?Voice: +39 050 315 3058 (op.2111)

Dmitri.
--


From riccardocorradini at yahoo.it  Fri Sep 11 09:16:08 2009
From: riccardocorradini at yahoo.it (Riccardo Corradini)
Date: Fri, 11 Sep 2009 14:16:08 +0000 (GMT)
Subject: Benchmarking Octave 322
In-Reply-To: <892b16670909110648w63864372mece42ba6769c2363@mail.gmail.com>
Message-ID: <289169.10228.qm@web25501.mail.ukl.yahoo.com>

I also have better results with custom optimized libraries
Speed of octave 3.2.2 on x86_64-unknown-linux-gnu
? Pentium II 350 MHz has score 1.0, higher numbers are better
Matrix inversion (LAPACK)??????? 34.47 +/- 21.0% (380 runs)
Schur decomposition (LAPACK)???? 29.56 +/- 1.3% (168 runs)
Differential equation (LSODE)??? 34.62 +/- 0.4% (819 runs)
Fourier transforms (FFTPACK)???? 13.09 +/- 0.2% (343 runs)
for loop???????????????????????? 54.78 +/- 0.6% (1008 runs)
-- Performance index (bm 2.3): 30

?cat /proc/cpuinfo
processor??? : 0
vendor_id??? : AuthenticAMD
cpu family??? : 15
model??? ??? : 107
model name??? : AMD Athlon(tm) 64 X2 Dual Core Processor 5600+

Bests
Riccardo
--- Ven 11/9/09, Dmitri A. Sergatskov <dasergatskov at gmail.com> ha scritto:

Da: Dmitri A. Sergatskov <dasergatskov at gmail.com>
Oggetto: Re: Benchmarking Octave 322
A: "Francesco Potorti`" <Potorti at isti.cnr.it>
Cc: help-octave at octave.org
Data: Venerd? 11 settembre 2009, 15:48

On Fri, Sep 11, 2009 at 6:48 AM, Francesco Potorti` <Potorti at isti.cnr.it> wrote:

> For a reference, here is what I see using my benchmark on an AMD Athlon
> 64 X2 3800+ running Debian's Octave 3.2.2:
>
> octave> benchmark
> Octave benchmark version bm 2.3
> Speed of octave 3.2.2 on x86_64-pc-linux-gnu
> ?Pentium II 350 MHz has score 1.0, higher numbers are better
> Matrix inversion (LAPACK) ? ? ? ? 9.05 +/- 0.3% (154 runs)
> Schur decomposition (LAPACK) ? ? 23.88 +/- 0.2% (126 runs)
> Differential equation (LSODE) ? ?22.83 +/- 0.3% (546 runs)
> Fourier transforms (FFTPACK) ? ? ?7.43 +/- 0.5% (196 runs)
> for loop ? ? ? ? ? ? ? ? ? ? ? ? 34.76 +/- 1.0% (658 runs)
> -- Performance index (bm 2.3): 17
>

It looks to me your ATLAS is not well optimized. Here is 3.2.2 on
Fedora 11 x86_64
(Opteron 180, at 2.4GHz):

Matrix inversion (LAPACK)? ? ? ? 31.47 +/- 2.6% (539 runs)
Schur decomposition (LAPACK)? ???39.68 +/- 0.1% (238 runs)
Differential equation (LSODE)? ? 26.19 +/- 0.6% (623 runs)
Fourier transforms (FFTPACK)? ? ? 7.09 +/- 1.9% (168 runs)
for loop? ? ? ? ? ? ? ? ? ? ? ???35.66 +/- 1.1% (665 runs)
-- Performance index (bm 2.3): 24


> --
> Francesco Potort? (ricercatore) ? ? ? ?Voice: +39 050 315 3058 (op.2111)

Dmitri.
--

_______________________________________________
Help-octave mailing list
Help-octave at octave.org
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/e8eeac22/attachment.html 

From Bard.Skaflestad at sintef.no  Fri Sep 11 09:04:30 2009
From: Bard.Skaflestad at sintef.no (=?iso-8859-1?Q?B=E5rd_Skaflestad?=)
Date: Fri, 11 Sep 2009 16:04:30 +0200
Subject: Using STR2FUNC in sub-function to refer to other sub-function(s)
Message-ID: <47AC1666471FFC41873637D50D111B812679568394@SINTEFEXMBX01.sintef.no>

All,

I'm experiencing a bit of bother using the STR2FUNC function.  For sake of argument I'll use the synthetic function 'foo' defined as

$ cat foo.m
   function h = foo()
      h = @sub1;
   end

   function h = sub1()
      h = str2func('sub2');
   end

   function h = sub2()
      h = @(x) sub3(x, 10);
   end

   function q = sub3(a, b)
      q = fprintf('sub3: (a,b) = (%f,%d)\n', a, b);
   end
$

In Octave I then get

--------------------8<----------------------------------------8<-------------------------

    octave:29> f = foo(), s1 = f(), s2 = s1(), s3 = s2(pi)
    f =

    sub1

    error: error creating function handle "@sub2"
    error: called from:
    error:   /home/bska/tmp/build/octave/3.2.2/foo.m at line 6, column 9

--------------------8<----------------------------------------8<-------------------------

while MATLAB yields the, to me, expected output

--------------------8<----------------------------------------8<-------------------------

    >> f = foo(), s1 = f(), s2 = s1(), s3 = s2(pi)

    f = 

        @sub1

    s1 = 

        @sub2

    s2 = 

        @(x)sub3(x,10)

    sub3: (a,b) = (3.141593,10)

    s3 =

        28

--------------------8<----------------------------------------8<-------------------------

My question is if this (failing to construct a function handle to 'sub2') is an error in Octave or if Octave's STR2FUNC implementation is not expected to have the ability to construct handles to (other) sub-functions of a "main" function.  I should note that if I replace the above 'sub1' function by the (equivalent, in this case)

    function h = sub1()
       h = @sub2;
    end

the above error goes away.  In the general case, however, I would really like the ability to use STR2FUNC here because the function name is not generally known until runtime in code.  Admittedly, my code uses only a bounded set of possible string values (roughly 5-10 alternatives), but I'd rather not insert a big SWITCH statement in there.  Any and all suggestions will be greatly appreciated.


Best regards,
-- 
B?rd Skaflestad                            <Bard.Skaflestad at sintef.no>
SINTEF ICT, Applied mathematics
Forskningsveien 1, No-0314 Oslo
Tel.: (+47) 2206 7642


For the record, I have

--------------------8<----------------------------------------8<-------------------------

    octave:26> ver
    ----------------------------------------------------------------------
    GNU Octave Version 3.2.2
    GNU Octave License: GNU General Public License
    Operating System: Linux 2.6.22.14-72.fc6 #1 SMP Wed Nov 21 14:10:25 EST 2007 x86_64
    ----------------------------------------------------------------------
    no packages installed.
    octave:27> 

--------------------8<----------------------------------------8<-------------------------

The interpreter is freshly built today (Friday the 11th of September 2009), and configured as

--------------------8<----------------------------------------8<-------------------------

/home/bska/tmp/downloads/octave-3.2.2/configure --enable-shared --enable-dl --prefix='/home/bska/system/usr/local' --libdir='/home/bska/system/usr/local/lib64' CXX='g++43' CC='gcc43' F77='gfortran43' FC='gfortran43' CPPFLAGS='-I/home/bska/system/usr/local/include -I/home/bska/system/usr/local/include/SuiteSparse' LDFLAGS='-L/home/bska/system/usr/local/lib64 -L/usr/local/lib64 -L/usr/lib64 -L/lib64' LIBS='-lmetis'

--------------------8<----------------------------------------8<-------------------------

whence 'configure' reports

--------------------8<----------------------------------------8<-------------------------

Octave is now configured for x86_64-unknown-linux-gnu

  Source directory:     /home/bska/tmp/downloads/octave-3.2.2
  Installation prefix:  /home/bska/system/usr/local
  C compiler:           gcc43   -Wall -W -Wshadow -Wformat -g -O2
  C++ compiler:         g++43   -Wall -W -Wshadow -Wold-style-cast -Wformat -g -O2
  Fortran compiler:     gfortran43 -O
  Fortran libraries:     -L/home/bska/system/usr/local/lib64 -L/usr/local/lib64 -L/usr/lib64 -L/lib64 -L/usr/local/lib/gcc/x86_64-unknown-linux-gnu/4.3.3 -L/usr/local/lib/gcc/x86_64-unknown-linux-gnu/4.3.3/../../../../lib64 -
L/lib/../lib64 -L/usr/lib/../lib64 -L/usr/local/lib/gcc/x86_64-unknown-linux-gnu/4.3.3/../../.. -lhdf5 -lz -lgfortranbegin -lgfortran -lm -lmetis
  BLAS libraries:       -llapack -lblas
  FFTW libraries:       -lfftw3 -lfftw3f
  GLPK libraries:       
  UMFPACK libraries:    -lumfpack
  AMD libraries:        -lamd
  CAMD libraries:       -lcamd
  COLAMD libraries:     -lcolamd
  CCOLAMD libraries:    -lccolamd
  CHOLMOD libraries:    -lcholmod
  CXSPARSE libraries:   -lcxsparse
  ARPACK libraries:     -larpack
  QRUPDATE libraries:   -lqrupdate
  HDF5 libraries:       -lhdf5
  CURL libraries:       
  REGEX libraries:      -L/home/bska/system/usr/local/lib64 -lpcre
  QHULL libraries:      -lqhull
  OPENGL libraries:     
  FLTK backend libs:    
  X11 include flags:    
  X11 libraries:        
  CARBON libraries:     
  LIBS:                 -lreadline  -lncurses -ldl -lblas -lhdf5 -lz -lm -lmetis
  Default pager:        less
  gnuplot:              gnuplot
  Magick config:        

  Do internal array bounds checking:  false
  Build static libraries:             false
  Build shared libraries:             true
  Dynamic Linking:                    true (dlopen)
  Include support for GNU readline:   true
  64-bit array dims and indexing:     false

configure:34470: WARNING: I didn't find gperf, but it's only a problem if you need to reconstruct oct-gperf.h
configure:34560: WARNING: GLPK library not found.  The glpk function for solving linear programs will be disabled.
configure:34566: WARNING: GraphicsMagick++ config script not found.  Assuming GraphicsMagic++ library and header files are missing, so imread will not be fully functional
configure:34614: WARNING: I didn't find texi2dvi, but it's only a problem if you need to reconstruct the DVI version of the manual
configure:34701: WARNING: "FLTK config script not found.  Native graphics will be disabled."
configure:34717: WARNING: 
configure:34719: WARNING: I didn't find the necessary libraries to compile native
configure:34721: WARNING: graphics. It isn't necessary to have native graphics
configure:34723: WARNING: but you will have to use gnuplot or you won't be able
configure:34725: WARNING: to use any of Octave's plotting commands
configure:34727: WARNING: 
configure:34737:

NOTE: libraries may be skipped if a library is not found OR
      if the library on your system is missing required features.

--------------------8<----------------------------------------8<-------------------------


From martin.ecarnot at free.fr  Fri Sep 11 07:08:55 2009
From: martin.ecarnot at free.fr (Martin)
Date: Fri, 11 Sep 2009 14:08:55 +0200
Subject: Plot - Remove points
Message-ID: <1252670935.4aaa3dd7014ef@imp.free.fr>

Hi,
I'm trying to plot several series of points on a same figure, and be able to
remove part of them.
I saw Matlab can do it with:
h1=plot([1 1]);
set(h1,'visible','off');

Is it possible to do something like that with Octave?
It seems 'visible' property is only available for a figure object, not for a
plot object.

Thank you

Martin

From Pascal.Dupuis at worldonline.be  Fri Sep 11 04:55:14 2009
From: Pascal.Dupuis at worldonline.be (Dupuis)
Date: Fri, 11 Sep 2009 02:55:14 -0700 (PDT)
Subject: non-linear programming - constraints
Message-ID: <25398227.post@talk.nabble.com>


Hello,

I'm trying to solve a non-linear minimisation problems where elements to
determine are part of a matrix: M = [a b; b c], with a and c >0. The
objective function requires that M be positive definite, i.e. ac - b^2 >=0.
I introduced an inequality constraint in sqp, yet during the search in
phi_L1, x sometimes is not acceptable with respect to the inequality
constraint. Is there a mean to strictly enforce the inequality constraint to
be valid ? Another way would be to take beff = sign(b)*(min(abs(b),
sqrt(a*c)) but I fear to introduce discontinuities in the inequality
function. Any hint ?
Regards

Pascal 
-- 
View this message in context: http://www.nabble.com/non-linear-programming---constraints-tp25398227p25398227.html
Sent from the Octave - General mailing list archive at Nabble.com.


From Potorti at isti.cnr.it  Fri Sep 11 09:59:33 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Fri, 11 Sep 2009 16:59:33 +0200
Subject: Benchmarking Octave 322
In-Reply-To: <69d8d540909110538u296bc919neede8a1f9252f927@mail.gmail.com>
References: <200909111111.57102.ct529@york.ac.uk>
	<69d8d540909110321s19e6c6bbg6c9bfcc21370b484@mail.gmail.com>
	<E1Mm4c7-0005nb-Fg@tucano.isti.cnr.it>
	<69d8d540909110538u296bc919neede8a1f9252f927@mail.gmail.com>
Message-ID: <E1Mm7b3-0001fh-Ou@tucano.isti.cnr.it>

>The package is also low priority for me; besides, it seems almost
>nobody uses it.

It's not user friendly.  While the disadvantege of my old package is
that it is not so modular as yours.  That is another reason why I always
thought about merging.  If someone wanted to do that...

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From bpabbott at mac.com  Fri Sep 11 10:29:32 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Fri, 11 Sep 2009 11:29:32 -0400
Subject: Plot - Remove points
In-Reply-To: <1252670935.4aaa3dd7014ef@imp.free.fr>
References: <1252670935.4aaa3dd7014ef@imp.free.fr>
Message-ID: <31127779484647633785705558090964518241-Webmail@me.com>

On Friday, September 11, 2009, at 08:08AM, "Martin" <martin.ecarnot at free.fr> wrote:
>Hi,
>I'm trying to plot several series of points on a same figure, and be able to
>remove part of them.
>I saw Matlab can do it with:
>h1=plot([1 1]);
>set(h1,'visible','off');
>
>Is it possible to do something like that with Octave?
>It seems 'visible' property is only available for a figure object, not for a
>plot object.
>
>Thank you
>
>Martin

That works for Octave 3.2.2. If you are running an older version, I recommend an upgrade.

Ben


From dfslezak at dc.uba.ar  Fri Sep 11 10:46:53 2009
From: dfslezak at dc.uba.ar (=?ISO-8859-1?Q?Diego_Fern=E1ndez_Slezak?=)
Date: Fri, 11 Sep 2009 11:46:53 -0400
Subject: Adding packages
Message-ID: <cfd0f9eb0909110846v12f96e4fs5022e056be659c13@mail.gmail.com>

Hello,

I have Octave 3.2.2 compiled for BlueGene/P and I want to add a package.
As I'm in a cross-compiling scenario I cannot run Octave and the "pkg
install ..."
I've tried to copy the .m files included in the package to the
share/packages directory, and adding the package dir to the path.
But it fails because it needs a .oct file.

How could I generate that oct in the cross-compiling situation and
where do I have to place it?

Thanks
Diego

From ct529 at york.ac.uk  Fri Sep 11 11:00:38 2009
From: ct529 at york.ac.uk (Corrado)
Date: Fri, 11 Sep 2009 17:00:38 +0100
Subject: Benchmarking Octave 322
In-Reply-To: <E1Mm7b3-0001fh-Ou@tucano.isti.cnr.it>
References: <200909111111.57102.ct529@york.ac.uk>
	<69d8d540909110538u296bc919neede8a1f9252f927@mail.gmail.com>
	<E1Mm7b3-0001fh-Ou@tucano.isti.cnr.it>
Message-ID: <200909111700.39331.ct529@york.ac.uk>

How is that for a laptop?

octave:1> benchmark
Octave benchmark version bm 2.3
Speed of octave 3.2.2 on x86_64-unknown-linux-gnu
  Pentium II 350 MHz has score 1.0, higher numbers are better
Matrix inversion (LAPACK)        14.98 +/- 0.1% (252 runs)
Schur decomposition (LAPACK)     54.52 +/- 0.4% (322 runs)
Differential equation (LSODE)    40.81 +/- 0.2% (980 runs)
Fourier transforms (FFTPACK)     23.09 +/- 0.4% (609 runs)
for loop                         75.20 +/- 3.6% (1365 runs)
-- Performance index (bm 2.3): 36

octave:2>

I think I could still optimise Matrix inversion though

On Friday 11 September 2009 15:59:33 Francesco Potorti` wrote:
> >The package is also low priority for me; besides, it seems almost
> >nobody uses it.
>
> It's not user friendly.  While the disadvantege of my old package is
> that it is not so modular as yours.  That is another reason why I always
> thought about merging.  If someone wanted to do that...



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From rob at jlab.org  Fri Sep 11 09:44:13 2009
From: rob at jlab.org (Rob Mahurin)
Date: Fri, 11 Sep 2009 10:44:13 -0400
Subject: Plot - Remove points
In-Reply-To: <1252670935.4aaa3dd7014ef@imp.free.fr>
References: <1252670935.4aaa3dd7014ef@imp.free.fr>
Message-ID: <20090911144413.GA18250@dyn21924539.dz.ornl.gov>

On Fri, Sep 11, 2009 at 02:08:55PM +0200, Martin wrote:
> I'm trying to plot several series of points on a same figure, and be able to
> remove part of them.
> I saw Matlab can do it with:
> h1=plot([1 1]);
> set(h1,'visible','off');
> 
> Is it possible to do something like that with Octave?
> It seems 'visible' property is only available for a figure object, not for a
> plot object.

Your example works for me in Octave 3.2.2.

Cheers,
Rob

From sraevsky at gmail.com  Fri Sep 11 12:15:40 2009
From: sraevsky at gmail.com (Scott Raevsky)
Date: Fri, 11 Sep 2009 10:15:40 -0700
Subject: freqz plot question
Message-ID: <b6a9929f0909111015xcebbf82gdc685aeb0988922b@mail.gmail.com>

Hello,

It's been years since I used Matlab, but one thing I remember is freqz
plotting the mag and phase response.  In octave, freqz plots a pass
band, a stop band, and phase.  What is the stop band for, and why is
it seperate from the pass band?  Seems like it should be just mag and
phase, where the pass and stop bands are part of the mag plot.

Thanks

-- 
Sent from my mobile device

From christy1865 at yahoo.com  Fri Sep 11 13:19:28 2009
From: christy1865 at yahoo.com (Kristen Richter)
Date: Fri, 11 Sep 2009 11:19:28 -0700 (PDT)
Subject: avifile (and all that jazz)
Message-ID: <887085.21640.qm@web30205.mail.mud.yahoo.com>

Dear helpful mailing list,

I am attempting to turn a series of plots into an avifile.  While I've installed the video add-on package from Octave-forge, I can't add frames to my .avi because the AVHandler can't seem to find the appropriate codecs.  Using the example below, found here: http://neacm.fe.up.pt/pub/octave/MAILING-LISTS/help-octave/2005/3694

# Open the AVI file
m = avifile("test.avi", "codec", "msmpeg4v2")

# Generate and add frames
for i = 1:100
  I = zeros(100,100);
  I(i,:) = i;
  I(:,i) = 200-i;
  addframe(m, I/255)
  printf(".")
endfor
printf("\n")

# Close the file
clear mI get the following error:

AVHandler: cannot find codec msmpeg4v2
error: avifile: AVHandler setup failed
addframe: error adding frame to avifile
I've tried the example, specifying many different codecs, including cinepak, indeo5, and indeo3.  I've checked my synaptic package manager for appropriate codecs... currently I've installed libavifile (vs. 0.7), which mentions the indeo library by name.

Are the codecs supposed to be installed in a specific directory? Has anyone come across similar issues in the past?
The example is pretty old... dated 2005, and I know the video package was updated last this past May.

Any insight on the situation would be greatly appreciated. As always, your help in the past has been invaluable.

-Kristen Richter

By the way, does anyone know of a good work around for the getframe function?... I have yet to think of a good way to capture the figure on the screen into an appropriately formatted image matrix.


      

From manishdkulkarni at gmail.com  Fri Sep 11 15:03:26 2009
From: manishdkulkarni at gmail.com (Manish Kulkarni)
Date: Fri, 11 Sep 2009 13:03:26 -0700
Subject: imread error
In-Reply-To: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>
References: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>
Message-ID: <8feb35460909111303g2ce1337ckd0acb2edca24b4c5@mail.gmail.com>

On Fri, Sep 11, 2009 at 12:57 PM, Manish Kulkarni <manishdkulkarni at gmail.com
> wrote:

> Dear everyone,
>
> I'm getting the following error with imread
>
> > tftimage = imread("Waterfall.jpg");
> error: imread: invalid image file: Magick++ exception: Magick: Unable to
> access configuration file (magic.mgk) reported
> by
> d:/files/admin/octaveforge_svn/trunk/octave-forge/admin/Windows/mingw32/graphicsmagick/GraphicsMagick-1.3.3/magick/bl
> ob.c:1824 (GetConfigureBlob)
> error: called from:
> error:   C:\Octave\3.2.0_gcc-4.3.0\share\octave\3.2.0\m\image\imread.m at
> line 74, column 7
> octave-3.2.0.exe:12:c:\Users\laptopGuru\Documents\mSilica\tftOctave
>
> would anyone advise how to fix this?
>
> Thanks a lot.
> Manish
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/709c95ec/attachment.html 

From wilson.rosa at gmail.com  Fri Sep 11 15:56:25 2009
From: wilson.rosa at gmail.com (Wilson de Oliveira)
Date: Fri, 11 Sep 2009 17:56:25 -0300
Subject: Netlab on Octave
In-Reply-To: <20090911001904.22946.qmail@web3802.mail.bbt.yahoo.co.jp>
References: <91e5dab20909100829r19723a32teb8d9f19c151b0f3@mail.gmail.com> 
	<20090911001904.22946.qmail@web3802.mail.bbt.yahoo.co.jp>
Message-ID: <91e5dab20909111356r7eaa6b8frea30f79f91b342c1@mail.gmail.com>

Tatsuro, great!  Thanks for your effort! Much appreciated!

After your help with the Octave right version and discovering that the real
problem was with the *uicontroI* function which creates user interface
control, I  went google?ing and found out that the subject of importing gui
related MATLAB functions is a trick one; particularly those defining an user
interface, making wide use of commands like uicontrol and uimenu.  I found
the sourceforge project  Octawrap <http://octawrap.sourceforge.net/> which I
am about to try which aims at resolving this incompatibility.

I am also going to try later the Octave extra
package<http://octave.sourceforge.net/packages.html> Java
Graphic Handles <http://octave.sourceforge.net/doc/funref_jhandles.html> at
 sourceforge.


Many thanks for help and patient.
Best regards

Wilson de Oliveira
UFRPE-DEInfo
http://www.cin.ufpe.br/~wrdo


2009/9/10 Tatsuro MATSUOKA <tmacchant at yahoo.co.jp>

> Hello Wilson de Oliveira
>
> --- Wilson de Oliveira wrote:
>
> > Ok. Sorry.
> >  For a start, Netlab is a Matlab toolbox for Pattern Recognition which
> > includes amongst others models Neural Networks. It is freely available
> at:
> > http://www.ncrg.aston.ac.uk/netlab/index.php
> > demnlab.m is a demonstration program which runs several models in gui
> > interface.
> >
> > As regard the version of the octave, thanks for your help! The octave
> 3.0.0
> > came with qtoctave - a gui interface for octave; Now that I?ve installed
> the
> > 3.2.2 version available at sourceforge.net I got fewer errors:
> >
> >  > demnlab
> > error: `uicontrol' undefined near line 33 column 3
> > error: called from:
> > error:   C:\Octave\Netlab\demnlab.m at line 33, column 3
> > octave-3.2.2.exe:6:C:\Octave\Netlab
> > >
> >
> > Thanks for your help
> > Wilson de Oliveira
> > UFRPE-DEInfo
> > http://www.cin.ufpe.br/~wrdo
> >
> >
>
> Unfortunately Octave does not support GUI graphic control.
> You should control graphics with graphic handle functions.
> Please  see section 15 "Plotting" in octave manual.
> You should read care fully "15.1 Plotting Basics" band "15.2 Advanced
> Plotting".
> Perhaps these types functions are compatible to MATLAB.
>
>
> I have tried a simple example in web page.
>
> ndata=20
> % Generate the matrix of inputs x and targets t.
> x = [0:1/19:1]';
> t = sin(2*pi*x) + 0.2*randn(ndata, 1);
>
> % Set up network parameters.
> net = mlp(1, 3, 1, 'linear')
>
> net =
> {
>  type = mlp
>  nin =  1
>  nhidden =  3
>  nout =  1
>  nwts =  10
>  outfn = linear
>  w1 =
>
>    -1.148633   1.104356   0.049478
>
>  b1 =
>
>    -0.070054   0.902782   1.349561
>
>  w2 =
>
>     0.17895
>     0.27658
>     0.36492
>
>  b2 =  0.66142
> }
> ****************
> % Set up vector of options for the optimiser.
> options = zeros(1,18);
> options(1) = 1;
> options(9) = 1;
> options(14) = 100;
>
> % Train using scaled conjugate gradients.
> [net, options] = netopt(net, options, x, t, 'scg');
> error: `fcnchk' undefined near line 68 column 5
> error: called from:
> error:   D:\usr\Tatsu\program\Octave\netlab\scg.m at line 68, column 3
> error:   D:\usr\Tatsu\program\Octave\netlab\netopt.m at line 36, column 17
>
> Octave does not support `fcnchk'
> But I found
>
> http://en.wikibooks.org/wiki/MATLAB_Programming/Differences_between_Octave_and_MATLAB
>
> ********
> If something (like Netlab) need a function called fcnchk you can just put
> the following into a file
> called fcnchk.m and put it somewhere Octave can find it:
> function f=fcnchk(x, n)
>  f = x;
> end
> ********
> I placed fcnchk.m under the same folder of netlab
>
> SO I get
>
> Checking gradient ...
>
>   analytic   diffs     delta
>
>  1.1904e+000  1.1904e+000  -1.3890e-009
>  4.8963e-001  4.8963e-001  8.2556e-010
>  1.1238e+000  1.1238e+000  -5.8918e-010
>  2.3292e+000  2.3292e+000  -1.3039e-010
>  1.1362e+000  1.1362e+000  -1.2848e-009
>  1.8523e+000  1.8523e+000  6.3412e-010
>  -1.3883e+001  -1.3883e+001  -1.9210e-010
>  2.0438e+001  2.0438e+001  5.2320e-010
>  1.9960e+001  1.9960e+001  -9.0928e-011
>  2.2659e+001  2.2659e+001  -8.4377e-010
> Cycle    1  Error    4.629483  Scale 1.000000e+000
> Cycle    2  Error    4.629483  Scale 5.000000e-001
> Cycle    3  Error    4.629483  Scale 2.000000e+000
> Cycle    4  Error    4.259136  Scale 8.000000e+000
> Cycle    5  Error    4.050134  Scale 4.000000e+000
> Cycle    6  Error    3.775366  Scale 2.000000e+000
> Cycle    7  Error    3.390233  Scale 2.000000e+000
> Cycle    8  Error    3.288238  Scale 1.000000e+000
> Cycle    9  Error    3.286530  Scale 5.000000e-001
> Cycle   10  Error    3.235898  Scale 2.500000e-001
> Cycle   11  Error    3.212532  Scale 1.250000e-001
> Cycle   12  Error    3.192542  Scale 6.250000e-002
> Cycle   13  Error    3.187294  Scale 3.125000e-002
>  :
>  :
> *****************************************
> % Plot the trained network predictions.
> plotvals = [0:0.01:1]';
> y = mlpfwd(net, plotvals);
> plot(plotvals, y, 'ob')
>
> gives a plot like attatchment files.
>
> ********
> Honestly I am not familiar with this type of work.
> In addition, I have no experience of MATLAB
>
> Hopefully users who use both MATLAB and octave give you much better
> suggestions.
>
> The aboves are all what I can do for you.
>
> Good luck!!
>
> Regards
>
> Tatsuro
>
>
> --------------------------------------
> Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
> http://pr.mail.yahoo.co.jp/ec10years/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/0d0199bd/attachment.html 

From romain.cotte at usherbrooke.ca  Fri Sep 11 16:07:51 2009
From: romain.cotte at usherbrooke.ca (R.C.)
Date: Fri, 11 Sep 2009 14:07:51 -0700 (PDT)
Subject: Wrapper fortran
Message-ID: <25408339.post@talk.nabble.com>


Hello,


I followed example  from http://wiki.octave.org/wiki.pl?OctaveFortran but I
have some trouble with my fortran subroutine.
Compilation works but when i call my function it gives ;

[f, xb] = wrapper (2, 4, rand(4,1), 1)
*** glibc detected *** octave: free(): invalid next size (fast): 0x09c20450
***
======= Backtrace: =========
/lib/tls/i686/cmov/libc.so.6[0xb5d31604]
/lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb5d335b6]
/usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0xb5eed231]
/usr/lib/libstdc++.so.6(_ZdaPv+0x1d)[0xb5eed28d]
/usr/lib/octave-3.0.1/liboctinterp.so(_ZN5ArrayIdED2Ev+0xc6)[0xb7a30c66]
etc...


My wrapper
***********************************************************************************
#include <octave/oct.h>
#include "f77-fcn.h"
extern "C"
  {
    int F77_FUNC (rosenf_b, ROSENF_b) (const int& ind, const int& n,
				const double* x,
				double* xb, double& f,
				const double& fb,
				double& g, double& ti, double& tr, double& td );
  }
  DEFUN_DLD (wrapper, args, ,
	    "- Loadable Function: [f, xb] = wrapper (ind, n, x, fb)\n\
  \n\
  Returns the f,xb.")
  {
    octave_value_list retval;
    Matrix X;
    
    const int IND = args(0).int_value(), N = args(1).int_value();
    const double FB;
    double G;
    NDArray XB;
    double F,TI,TR,TD;
    double *av;
    
     FB = args(3).double_value();
     X = args(2).matrix_value();
     av = XB.fortran_vec();
    

    F77_XFCN (rosenf_b, ROSENF_B,
	    (IND, N,
	    X.fortran_vec(),
	    av,
	    F,
	    FB,
	    G,
	    TI, TR, TD) );
    if (f77_exception_encountered)
      {
	error ("unrecoverable error in rosenf_b");
	return retval;
      }
    retval(0) = octave_value (F);
    retval(1) = octave_value(XB);
    
    return retval;
  }
*************************************************************************




Fortran function
************************************************************************
      SUBROUTINE ROSENF_B(ind, n, x, xb, f, fb, g, ti, tr, td)
      IMPLICIT NONE
      INTEGER ind, n, ti(*)
      DOUBLE PRECISION x(n), f, g(n), td(*)
      DOUBLE PRECISION xb(n), fb
      REAL tr(*)
C
      DOUBLE PRECISION y, p
      INTEGER i
      DOUBLE PRECISION tempb
      INTEGER ii1
      p = 100.0d0
      IF (ind .EQ. 2 .OR. ind .EQ. 4) THEN
        DO ii1=1,n
          xb(ii1) = 0.D0
        ENDDO
        DO i=n,2,-1
          tempb = p*2*(x(i)-x(i-1)**2)*fb
          xb(i) = xb(i) + tempb - 2*(1.0d0-x(i))*fb
          xb(i-1) = xb(i-1) - 2*x(i-1)*tempb
        ENDDO
        fb = 0.D0
      ELSE
        DO ii1=1,n
          xb(ii1) = 0.D0
        ENDDO
      END IF
      END
**********************************************************************



What do you suggest ?
Thanks

-- 
View this message in context: http://www.nabble.com/Wrapper-fortran-tp25408339p25408339.html
Sent from the Octave - General mailing list archive at Nabble.com.


From manishdkulkarni at gmail.com  Fri Sep 11 14:57:11 2009
From: manishdkulkarni at gmail.com (Manish Kulkarni)
Date: Fri, 11 Sep 2009 12:57:11 -0700
Subject: imread error
Message-ID: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>

 Dear everyone,

I'm getting the following error with imread

> tftimage = imread("Waterfall.jpg");
error: imread: invalid image file: Magick++ exception: Magick: Unable to
access configuration file (magic.mgk) reported
by
d:/files/admin/octaveforge_svn/trunk/octave-forge/admin/Windows/mingw32/graphicsmagick/GraphicsMagick-1.3.3/magick/bl
ob.c:1824 (GetConfigureBlob)
error: called from:
error:   C:\Octave\3.2.0_gcc-4.3.0\share\octave\3.2.0\m\image\imread.m at
line 74, column 7
octave-3.2.0.exe:12:c:\Users\laptopGuru\Documents\mSilica\tftOctave

would anyone advise how to fix this?

Thanks a lot.
Manish
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090911/ae8c30bb/attachment-0001.html 

From bharat at arithos.com  Fri Sep 11 17:23:22 2009
From: bharat at arithos.com (bharat pathak)
Date: Sat, 12 Sep 2009 03:53:22 +0530
Subject: freqz plot question
References: <b6a9929f0909111015xcebbf82gdc685aeb0988922b@mail.gmail.com>
Message-ID: <4687E502E05B47299150BD8C2426D928@bharatPC>

Dear Scott,

when we design any digital filter, example low pass, high pass,
band pass and band stop. the filter is supposed to pass certain
frequencies to the output and stop certain other frequencies.

practically we cannot stop the frequencies, we just attenuate
them enough such that they do not fall in the dynamic range
of the datapath. for example if your datapath is 8 bit (DR = 40 dB)
then if you provide 40dB attenuation (1/100 in linear) to incoming
signal of full strength then we would not see such signal at the
output, provided the output datapath is also having 8 bit precision.

hence stop band is the region in frequency domain magnitude response 
plot where those input signals whose frequency lies in the stop band 
region would undergo certain attenuation before coming out of the
processing module.

hope this gives some clarity,

Regards
Bharat Pathak

Arithos Designs
www.Arithos.com

DSP Design Consultancy and Technical Trainings.

--------------------------------------------------
From: "Scott Raevsky" <sraevsky at gmail.com>
Sent: Friday, September 11, 2009 10:45 PM
To: <help-octave at octave.org>
Subject: freqz plot question

> Hello,
> 
> It's been years since I used Matlab, but one thing I remember is freqz
> plotting the mag and phase response.  In octave, freqz plots a pass
> band, a stop band, and phase.  What is the stop band for, and why is
> it seperate from the pass band?  Seems like it should be just mag and
> phase, where the pass and stop bands are part of the mag plot.
> 
> Thanks
> 
> -- 
> Sent from my mobile device
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave 

From sraevsky at gmail.com  Fri Sep 11 17:34:51 2009
From: sraevsky at gmail.com (Scott Raevsky)
Date: Fri, 11 Sep 2009 15:34:51 -0700
Subject: freqz plot question
In-Reply-To: <4687E502E05B47299150BD8C2426D928@bharatPC>
References: <b6a9929f0909111015xcebbf82gdc685aeb0988922b@mail.gmail.com>
	<4687E502E05B47299150BD8C2426D928@bharatPC>
Message-ID: <b6a9929f0909111534ydd59aey6a2de6e8906383d2@mail.gmail.com>

Thanks.  I should clarify that I know these concepts well.  However,
whenever I have seen frequency plots before, they include magnitude
and phase--two separate plots.  The "pass band" and "stop band" (if
they truly exist) are plotted on the magnitude plot.  But freqz in
octave plots two separate plots for "pass band" and "stop band," which
makes no sense to me.  Why does it do this?

On 9/11/09, bharat pathak <bharat at arithos.com> wrote:
> Dear Scott,
>
> when we design any digital filter, example low pass, high pass,
> band pass and band stop. the filter is supposed to pass certain
> frequencies to the output and stop certain other frequencies.
>
> practically we cannot stop the frequencies, we just attenuate
> them enough such that they do not fall in the dynamic range
> of the datapath. for example if your datapath is 8 bit (DR = 40 dB)
> then if you provide 40dB attenuation (1/100 in linear) to incoming
> signal of full strength then we would not see such signal at the
> output, provided the output datapath is also having 8 bit precision.
>
> hence stop band is the region in frequency domain magnitude response
> plot where those input signals whose frequency lies in the stop band
> region would undergo certain attenuation before coming out of the
> processing module.
>
> hope this gives some clarity,
>
> Regards
> Bharat Pathak
>
> Arithos Designs
> www.Arithos.com
>
> DSP Design Consultancy and Technical Trainings.
>
> --------------------------------------------------
> From: "Scott Raevsky" <sraevsky at gmail.com>
> Sent: Friday, September 11, 2009 10:45 PM
> To: <help-octave at octave.org>
> Subject: freqz plot question
>
>> Hello,
>>
>> It's been years since I used Matlab, but one thing I remember is freqz
>> plotting the mag and phase response.  In octave, freqz plots a pass
>> band, a stop band, and phase.  What is the stop band for, and why is
>> it seperate from the pass band?  Seems like it should be just mag and
>> phase, where the pass and stop bands are part of the mag plot.
>>
>> Thanks
>>
>> --
>> Sent from my mobile device
>> _______________________________________________
>> Help-octave mailing list
>> Help-octave at octave.org
>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>

-- 
Sent from my mobile device

From tmacchant at yahoo.co.jp  Fri Sep 11 18:02:33 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Sat, 12 Sep 2009 08:02:33 +0900 (JST)
Subject: Netlab on Octave
In-Reply-To: <91e5dab20909111356r7eaa6b8frea30f79f91b342c1@mail.gmail.com>
Message-ID: <20090911230233.44339.qmail@web3815.mail.bbt.yahoo.co.jp>

Hello


--- Wilson de Oliveira wrote:

> I am also going to try later the Octave extra
> package<http://octave.sourceforge.net/packages.html> Java
> Graphic Handles <http://octave.sourceforge.net/doc/funref_jhandles.html> at
>  sourceforge.

If you try to install jhandle, please see suggestions by Michael:

http://www.nabble.com/Octave-3.2.0-mingw-jhandles-td24024734.html

Regards

Tatsuro

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From hinow at uwm.edu  Fri Sep 11 18:11:59 2009
From: hinow at uwm.edu (Peter Hinow)
Date: Fri, 11 Sep 2009 18:11:59 -0500 (CDT)
Subject: error with my octave installation?
In-Reply-To: <896547768.6004571252710623182.JavaMail.root@mail02.pantherlink.uwm.edu>
Message-ID: <1193948361.6004791252710719677.JavaMail.root@mail02.pantherlink.uwm.edu>

Hello all,
as I'm totally new to octave, I don't know yet how to report a problem or how to diagnose one. I did not install octave 3.0.1 explicitly, it came with the installation of Ubuntu 9.04 from a live CD. Then I encountered a problem using the pdist.m function. I got pdist.m and squareform.m from 
http://octave.sourceforge.net/doc/funref_statistics.html

The following set of commands works just fine
a=rand(257,3);
d1=squareform(pdist(a,"euclidean"));
However, increasing the size of the data by one
a=rand(258,3);
d1=squareform(pdist(a,"euclidean"));
gives a large number of error messages that are all

error: called from `nchoosek' in file `/usr/share/octave/3.0.1/m/specfun/nchoosek.m'
error: evaluating assignment expression near line 79, column 7
error: evaluating if command near line 69, column 3

How do I check whether my octave installation is correct?
Thank you and best regards,
Peter

From rabuter at eso.org  Sat Sep 12 06:07:18 2009
From: rabuter at eso.org (Roberto Abuter)
Date: Sat, 12 Sep 2009 13:07:18 +0200
Subject: interp1 -  bug 
Message-ID: <4AAB80E6.7040106@eso.org>

Hi :

   I am using the most basic form of interp1 in both matlab and octave.
   To my surprise interp1 in octave (3.0.2)  is producing some very 
strange results on my data.

   I have a set of two vectors X, Y where X has a sawtooth shape (4 
repetitions) and Y has 4 repetitions of an increasing/decreasing 
amplitude sine wave.
   to start very simple I just check the interp1 engine by making :
  
   XI = X;
   YI = interp1(X,Y,XI)


   to my surprise, the result YI  comes out not identical to Y .  
  
   well, if I try to interp1 with XI as my target grid ( a regular one 
) I get absolutely wierd results.

hope someone can give me a hint.

many thanks in advance,

Roberto Abuter
VLTI project-
E.S.O ( Europeam Southern Observatory)

  



From jwe at octave.org  Sat Sep 12 06:12:12 2009
From: jwe at octave.org (John W. Eaton)
Date: Sat, 12 Sep 2009 07:12:12 -0400
Subject: interp1 -  bug 
In-Reply-To: <4AAB80E6.7040106@eso.org>
References: <4AAB80E6.7040106@eso.org>
Message-ID: <19115.33292.209692.398085@segfault.lan>

On 12-Sep-2009, Roberto Abuter wrote:

|    I am using the most basic form of interp1 in both matlab and octave.
|    To my surprise interp1 in octave (3.0.2)  is producing some very 
| strange results on my data.
| 
|    I have a set of two vectors X, Y where X has a sawtooth shape (4 
| repetitions) and Y has 4 repetitions of an increasing/decreasing 
| amplitude sine wave.
|    to start very simple I just check the interp1 engine by making :
|   
|    XI = X;
|    YI = interp1(X,Y,XI)
| 
| 
|    to my surprise, the result YI  comes out not identical to Y .  
|   
|    well, if I try to interp1 with XI as my target grid ( a regular one 
| ) I get absolutely wierd results.
| 
| hope someone can give me a hint.

Is X monotonically increasing (or decreasing)?  From your description
it sounds like it is not.

If you think you've found a bug, then please include enough
information so that someone can reproduce precisely what you are
doing and report it to bug at octave.org.

In any case, there is no need to cross post to both the bug and help
lists.

jwe

From rabuter at eso.org  Sat Sep 12 06:41:37 2009
From: rabuter at eso.org (Roberto Abuter)
Date: Sat, 12 Sep 2009 13:41:37 +0200
Subject: interp1 -  bug
In-Reply-To: <19115.33292.209692.398085@segfault.lan>
References: <4AAB80E6.7040106@eso.org> <19115.33292.209692.398085@segfault.lan>
Message-ID: <4AAB88F1.8040608@eso.org>


I am sorry for the double posting, I am very new to this list.
Indeed, X is not monotonically increasing and I now see in the Octave 
function description that X has to be monotonic.
Sorry for missing that !
I guess it is just a difference in behavior between Matlab and Octave.
one can always reshuffle (X,Y) in such a way X becames monotonic.

many thanks again,
Roberto AButer

John W. Eaton wrote:

>On 12-Sep-2009, Roberto Abuter wrote:
>
>|    I am using the most basic form of interp1 in both matlab and octave.
>|    To my surprise interp1 in octave (3.0.2)  is producing some very 
>| strange results on my data.
>| 
>|    I have a set of two vectors X, Y where X has a sawtooth shape (4 
>| repetitions) and Y has 4 repetitions of an increasing/decreasing 
>| amplitude sine wave.
>|    to start very simple I just check the interp1 engine by making :
>|   
>|    XI = X;
>|    YI = interp1(X,Y,XI)
>| 
>| 
>|    to my surprise, the result YI  comes out not identical to Y .  
>|   
>|    well, if I try to interp1 with XI as my target grid ( a regular one 
>| ) I get absolutely wierd results.
>| 
>| hope someone can give me a hint.
>
>Is X monotonically increasing (or decreasing)?  From your description
>it sounds like it is not.
>
>If you think you've found a bug, then please include enough
>information so that someone can reproduce precisely what you are
>doing and report it to bug at octave.org.
>
>In any case, there is no need to cross post to both the bug and help
>lists.
>
>jwe
>  
>


From soren at hauberg.org  Sat Sep 12 06:48:28 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Sat, 12 Sep 2009 13:48:28 +0200
Subject: interp1 -  bug
In-Reply-To: <4AAB88F1.8040608@eso.org>
References: <4AAB80E6.7040106@eso.org>
	<19115.33292.209692.398085@segfault.lan>  <4AAB88F1.8040608@eso.org>
Message-ID: <1252756108.4464.62.camel@sh-t400>

l?r, 12 09 2009 kl. 13:41 +0200, skrev Roberto Abuter:
> I guess it is just a difference in behavior between Matlab and Octave.
> one can always reshuffle (X,Y) in such a way X becames monotonic.

Is there any particular reason why we are not doing this in 'interp1'?
Couldn't just add

  [x, idx] = sort (x);
  y = y (idx);

to 'interp1' ?

S?ren


From jwe at octave.org  Sat Sep 12 07:04:07 2009
From: jwe at octave.org (John W. Eaton)
Date: Sat, 12 Sep 2009 08:04:07 -0400
Subject: interp1 -  bug
In-Reply-To: <1252756108.4464.62.camel@sh-t400>
References: <4AAB80E6.7040106@eso.org> <19115.33292.209692.398085@segfault.lan>
	<4AAB88F1.8040608@eso.org> <1252756108.4464.62.camel@sh-t400>
Message-ID: <19115.36407.147312.466204@segfault.lan>

On 12-Sep-2009, S?ren Hauberg wrote:

| l?r, 12 09 2009 kl. 13:41 +0200, skrev Roberto Abuter:
| > I guess it is just a difference in behavior between Matlab and Octave.
| > one can always reshuffle (X,Y) in such a way X becames monotonic.
| 
| Is there any particular reason why we are not doing this in 'interp1'?
| Couldn't just add
| 
|   [x, idx] = sort (x);
|   y = y (idx);
| 
| to 'interp1' ?

Suppose I have a set of points that approximate a circle:

  theta = 0:0.2:2*pi;
  x = cos(theta);
  y = sin(theta);
  plot (x, y);

and I want to interpolate between two of these points.  If I leave the
points ordered as they are, I need to pick a range where the
interpolation will make sense.  If I just blindly sort the X and Y
values, I get something that doesn't look a lot like a circle, so
interpolating between two of the points is not likely to give me the
result I'm looking for:

  [x, idx] = sort (x);
  y = y (idx);
  plot (x, y);

jwe


From soren at hauberg.org  Sat Sep 12 07:15:48 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Sat, 12 Sep 2009 14:15:48 +0200
Subject: interp1 -  bug
In-Reply-To: <19115.36407.147312.466204@segfault.lan>
References: <4AAB80E6.7040106@eso.org>
	<19115.33292.209692.398085@segfault.lan> <4AAB88F1.8040608@eso.org>
	<1252756108.4464.62.camel@sh-t400>
	<19115.36407.147312.466204@segfault.lan>
Message-ID: <1252757748.4464.63.camel@sh-t400>

l?r, 12 09 2009 kl. 08:04 -0400, skrev John W. Eaton:
> Suppose I have a set of points that approximate a circle:
> 
>   theta = 0:0.2:2*pi;
>   x = cos(theta);
>   y = sin(theta);
>   plot (x, y);
> 
> and I want to interpolate between two of these points.  If I leave the
> points ordered as they are, I need to pick a range where the
> interpolation will make sense.  If I just blindly sort the X and Y
> values, I get something that doesn't look a lot like a circle, so
> interpolating between two of the points is not likely to give me the
> result I'm looking for:
> 
>   [x, idx] = sort (x);
>   y = y (idx);
>   plot (x, y);

Good point! I was only considering surjective data.

Thanks
S?ren


From highegg at gmail.com  Sat Sep 12 14:33:06 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Sat, 12 Sep 2009 21:33:06 +0200
Subject: interp1 - bug
In-Reply-To: <1252756108.4464.62.camel@sh-t400>
References: <4AAB80E6.7040106@eso.org> <19115.33292.209692.398085@segfault.lan>
	<4AAB88F1.8040608@eso.org> <1252756108.4464.62.camel@sh-t400>
Message-ID: <69d8d540909121233n58d58248o9f871d7a814a8c0@mail.gmail.com>

On Sat, Sep 12, 2009 at 1:48 PM, S?ren Hauberg <soren at hauberg.org> wrote:
> l?r, 12 09 2009 kl. 13:41 +0200, skrev Roberto Abuter:
>> I guess it is just a difference in behavior between Matlab and Octave.
>> one can always reshuffle (X,Y) in such a way X becames monotonic.
>
> Is there any particular reason why we are not doing this in 'interp1'?
> Couldn't just add
>
> ?[x, idx] = sort (x);
> ?y = y (idx);
>
> to 'interp1' ?
>
> S?ren
>

Another reason is performance. At least for the default linear
interpolation, interpolating a single point in a table of length N
costs O(log(N)). By always first sorting the table you would degrade
that to O(N). This is precisely why for a binary search, you normally
"trust" the table to be sorted - checking for that would often defeat
the purpose of binary search in the first place (not always, but
often).


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From smpearson at hotmail.com  Sat Sep 12 15:08:23 2009
From: smpearson at hotmail.com (mark8)
Date: Sat, 12 Sep 2009 13:08:23 -0700 (PDT)
Subject: Statically linking Octave to C++ program
Message-ID: <25417085.post@talk.nabble.com>


Hi,
I need to statically link the Octave library within a C++ program so that it
can be used on a distributed grid. For other libraries, I have made symlinks
with altered names to the static library (such as libgsl-s.a for the gsl
library) in the same library directory, and then compile with, for example,
-lgsl-s. 

This works fine for any other library except Octave. With Octave, I can
#include<octave/oct.h> and of course it links in fine so long as I don't
call any octave functions. As soon as I try to declare an octave object
(such as Matrix m(2,2);), I get hundreds of lines of "undefined symbols"
followed by 
ld: symbol(s) not found

Right now, I'm working with not much more than a "Hello World" program. It
is fine if I use -loctave but not if I use -loctave-s where liboctave-s.a is
a symlink to liboctave.a.
I've also tried to use the -static compiler flag
I do have the static libraries octave-3.2.2/liboctave.a (and libcruft.a and
liboctinterp.a)

Other details:
I'm using octave 3.2.2 on Mac OSX 10.5
Octave obtained via macports (but I've also built it myself with no
differences)
Compiler: gcc 4.0.1 (but other versions give the same result)

Is it possible to do this static linking? If so, can anyone suggest how?
Thank you very much,
Mark


-- 
View this message in context: http://www.nabble.com/Statically-linking-Octave-to-C%2B%2B-program-tp25417085p25417085.html
Sent from the Octave - General mailing list archive at Nabble.com.


From Scott.Raevsky at kmob.com  Sat Sep 12 15:03:23 2009
From: Scott.Raevsky at kmob.com (Scott.Raevsky)
Date: Sat, 12 Sep 2009 13:03:23 -0700
Subject: freqz question (clarified)
Message-ID: <8A83011125D82746BD74D09F439F4E2016AEE15F@oc-exchange-3.kmob.com>

Hello,
 
Just to clarify my previous question, Matlab's output of freqz is as
follows:
 
freqz(b,a,...) plots the magnitude and unwrapped phase of the frequency
response of the filter. The plot is displayed in the current figure
window.
 
Why does Octave's freqz not plot magnitude and phase?  Instead it plots
1) pass band, 2) stop band, and 3) phase.  Plotting a separate pass band
and stop band doesn't make sense to me.
 
Thanks
 

MAIL.KMOB.COM made the following annotations
---------------------------------------------------------------------
PRIVILEGED AND CONFIDENTIAL COMMUNICATION

This electronic transmission, and any documents attached
hereto, may contain confidential and/or legally privileged
information.  The information is intended only for use by
the recipient named above. If you have received this
electronic message in error, please notify the sender and
delete the electronic message. Any disclosure, copying,
distribution, or use of the contents of information
received in error is strictly prohibited.

---------------------------------------------------------------------

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090912/4c2ea81c/attachment.html 

From smpearson at hotmail.com  Sat Sep 12 13:40:13 2009
From: smpearson at hotmail.com (mark8)
Date: Sat, 12 Sep 2009 11:40:13 -0700 (PDT)
Subject: Statically linking Octave to C++ program
Message-ID: <25417085.post@talk.nabble.com>


Hi,
I need to statically link the Octave library within a C++ program so that it
can be used on a distributed grid. For other libraries, I have made symlinks
with altered names to the static library (such as libgsl-s.a for the gsl
library) in the same library directory, and then compile with, for example,
-lgsl-s. 

This works fine for any other library except Octave. With Octave, I can
#include<octave/oct.h> and of course it links in fine so long as I don't
call any octave functions. As soon as I try to declare an octave object
(such as Matrix m(2,2);), I get hundreds of lines of "undefined symbols"
followed by 
ld: symbol(s) not found

Right now, I'm working with not much more than a "Hello World" program. It
is fine if I use -loctave but not if I use -loctave-s where liboctave-s.a is
a symlink to liboctave.a.
I've also tried to use the -static compiler flag
I do have the static libraries octave-3.2.2/liboctave.a (and libcruft.a and
liboctinterp.a)

Other details:
I'm using octave 3.2.2 on Mac OSX 10.5
Octave obtained via macports (but I've also built it myself with no
differences)
Compiler: gcc 4.0.1 (but other versions give the same result)

Is it possible to do this static linking? If so, can anyone suggest how?
Thank you very much,
Mark


-- 
View this message in context: http://www.nabble.com/Statically-linking-Octave-to-C%2B%2B-program-tp25417085p25417085.html
Sent from the Octave - General mailing list archive at Nabble.com.


From dbateman at dbateman.org  Sat Sep 12 16:24:26 2009
From: dbateman at dbateman.org (David Bateman)
Date: Sat, 12 Sep 2009 23:24:26 +0200
Subject: Statically linking Octave to C++ program
In-Reply-To: <25417085.post@talk.nabble.com>
References: <25417085.post@talk.nabble.com>
Message-ID: <4AAC118A.9030403@dbateman.org>

mark8 wrote:
> Hi,
> I need to statically link the Octave library within a C++ program so that it
> can be used on a distributed grid. For other libraries, I have made symlinks
> with altered names to the static library (such as libgsl-s.a for the gsl
> library) in the same library directory, and then compile with, for example,
> -lgsl-s. 
>
> This works fine for any other library except Octave. With Octave, I can
> #include<octave/oct.h> and of course it links in fine so long as I don't
> call any octave functions. As soon as I try to declare an octave object
> (such as Matrix m(2,2);), I get hundreds of lines of "undefined symbols"
> followed by 
> ld: symbol(s) not found
>
> Right now, I'm working with not much more than a "Hello World" program. It
> is fine if I use -loctave but not if I use -loctave-s where liboctave-s.a is
> a symlink to liboctave.a.
> I've also tried to use the -static compiler flag
> I do have the static libraries octave-3.2.2/liboctave.a (and libcruft.a and
> liboctinterp.a)
>
> Other details:
> I'm using octave 3.2.2 on Mac OSX 10.5
> Octave obtained via macports (but I've also built it myself with no
> differences)
> Compiler: gcc 4.0.1 (but other versions give the same result)
>
> Is it possible to do this static linking? If so, can anyone suggest how?
> Thank you very much,
> Mark
>
>
>   
Try the "-static" flag to the octave configuration script.. The reason 
for your issues is that the octave oct-files are essentially libraries 
and these need to be linked in as well. Note that any octave-forge or 
home built oct-file that you will want to use will also have to be 
linked in... This gets messy pretty quickly

D.


From smpearson at hotmail.com  Sat Sep 12 17:02:13 2009
From: smpearson at hotmail.com (mark8)
Date: Sat, 12 Sep 2009 15:02:13 -0700 (PDT)
Subject: Statically linking Octave to C++ program
In-Reply-To: <4AAC118A.9030403@dbateman.org>
References: <25417085.post@talk.nabble.com> <4AAC118A.9030403@dbateman.org>
Message-ID: <25418627.post@talk.nabble.com>




dbateman wrote:
> 
> mark8 wrote:
>> Hi,
>> I need to statically link the Octave library within a C++ program so that
>> it
>> can be used on a distributed grid. For other libraries, I have made
>> symlinks
>> with altered names to the static library (such as libgsl-s.a for the gsl
>> library) in the same library directory, and then compile with, for
>> example,
>> -lgsl-s. 
>>
>> This works fine for any other library except Octave. With Octave, I can
>> #include<octave/oct.h> and of course it links in fine so long as I don't
>> call any octave functions. As soon as I try to declare an octave object
>> (such as Matrix m(2,2);), I get hundreds of lines of "undefined symbols"
>> followed by 
>> ld: symbol(s) not found
>>
>> Right now, I'm working with not much more than a "Hello World" program.
>> It
>> is fine if I use -loctave but not if I use -loctave-s where liboctave-s.a
>> is
>> a symlink to liboctave.a.
>> I've also tried to use the -static compiler flag
>> I do have the static libraries octave-3.2.2/liboctave.a (and libcruft.a
>> and
>> liboctinterp.a)
>>
>> Other details:
>> I'm using octave 3.2.2 on Mac OSX 10.5
>> Octave obtained via macports (but I've also built it myself with no
>> differences)
>> Compiler: gcc 4.0.1 (but other versions give the same result)
>>
>> Is it possible to do this static linking? If so, can anyone suggest how?
>> Thank you very much,
>> Mark
>>
>>
>>   
> Try the "-static" flag to the octave configuration script.. The reason 
> for your issues is that the octave oct-files are essentially libraries 
> and these need to be linked in as well. Note that any octave-forge or 
> home built oct-file that you will want to use will also have to be 
> linked in... This gets messy pretty quickly
> 
> D.
> 
> 
> 

Do you mean:
./configure --enable-static 
Doesn't the fact that I have the libraries liboctave.a etc. suggest that was
done already?

Or do you mean adding the --static flag to the Makefile? If so, where? I
tried adding it to the 'all' target so that it reads:
all: header-msg config-check
	$(MAKE) -f octMakefile all
but make complained.
-- 
View this message in context: http://www.nabble.com/Statically-linking-Octave-to-C%2B%2B-program-tp25417085p25418627.html
Sent from the Octave - General mailing list archive at Nabble.com.


From smpearson at hotmail.com  Sat Sep 12 22:21:06 2009
From: smpearson at hotmail.com (mark8)
Date: Sat, 12 Sep 2009 20:21:06 -0700 (PDT)
Subject: Statically linking Octave to C++ program
In-Reply-To: <25418627.post@talk.nabble.com>
References: <25417085.post@talk.nabble.com> <4AAC118A.9030403@dbateman.org>
	<25418627.post@talk.nabble.com>
Message-ID: <25420243.post@talk.nabble.com>




mark8 wrote:
> 
> 
> Do you mean:
> ./configure --enable-static 
> Doesn't the fact that I have the libraries liboctave.a etc. suggest that
> was done already?
> 
> Or do you mean adding the --static flag to the Makefile? If so, where? I
> tried adding it to the 'all' target so that it reads:
> all: header-msg config-check
> 	$(MAKE) -f octMakefile all
> but make complained.
> 

For what it's worth, I just tried 
./configure --enable-static
which produced liboctave.a. However, if I run  ranlib liboctave.a, I get:
> ranlib liboctave.a
ranlib: file: liboctave.a(oct-fftw.o) has no symbols
ranlib: file: liboctave.a(f2c-main.o) has no symbols
ranlib: file: liboctave.a(mkdir.o) has no symbols
ranlib: file: liboctave.a(rename.o) has no symbols
ranlib: file: liboctave.a(rmdir.o) has no symbols
ranlib: file: liboctave.a(strftime.o) has no symbols
ranlib: file: liboctave.a(strptime.o) has no symbols
ranlib: file: liboctave.a(strcasecmp.o) has no symbols
ranlib: file: liboctave.a(strncase.o) has no symbols
ranlib: file: liboctave.a(tempname.o) has no symbols
ranlib: file: liboctave.a(tempnam.o) has no symbols
ranlib: file: liboctave.a(oct-rl-edit.o) has no symbols
ranlib: file: liboctave.a(oct-rl-hist.o) has no symbols

feeling stumped...

-- 
View this message in context: http://www.nabble.com/Statically-linking-Octave-to-C%2B%2B-program-tp25417085p25420243.html
Sent from the Octave - General mailing list archive at Nabble.com.


From dbateman at dbateman.org  Sun Sep 13 04:55:07 2009
From: dbateman at dbateman.org (David Bateman)
Date: Sun, 13 Sep 2009 11:55:07 +0200
Subject: Statically linking Octave to C++ program
In-Reply-To: <25418627.post@talk.nabble.com>
References: <25417085.post@talk.nabble.com> <4AAC118A.9030403@dbateman.org>
	<25418627.post@talk.nabble.com>
Message-ID: <4AACC17B.3050708@dbateman.org>

mark8 wrote:
>   
> Do you mean:
> ./configure --enable-static 
> Doesn't the fact that I have the libraries liboctave.a etc. suggest that was
> done already?
>
>   
yes. You might need --disable-shared as well

D.



From dbateman at dbateman.org  Sun Sep 13 04:56:10 2009
From: dbateman at dbateman.org (David Bateman)
Date: Sun, 13 Sep 2009 11:56:10 +0200
Subject: Statically linking Octave to C++ program
In-Reply-To: <25420243.post@talk.nabble.com>
References: <25417085.post@talk.nabble.com>
	<4AAC118A.9030403@dbateman.org>	<25418627.post@talk.nabble.com>
	<25420243.post@talk.nabble.com>
Message-ID: <4AACC1BA.8050908@dbateman.org>

mark8 wrote:
>
> mark8 wrote:
>   
>> Do you mean:
>> ./configure --enable-static 
>> Doesn't the fact that I have the libraries liboctave.a etc. suggest that
>> was done already?
>>
>> Or do you mean adding the --static flag to the Makefile? If so, where? I
>> tried adding it to the 'all' target so that it reads:
>> all: header-msg config-check
>> 	$(MAKE) -f octMakefile all
>> but make complained.
>>
>>     
>
> For what it's worth, I just tried 
> ./configure --enable-static
> which produced liboctave.a. However, if I run  ranlib liboctave.a, I get:
>   
>> ranlib liboctave.a
>>     
> ranlib: file: liboctave.a(oct-fftw.o) has no symbols
> ranlib: file: liboctave.a(f2c-main.o) has no symbols
> ranlib: file: liboctave.a(mkdir.o) has no symbols
> ranlib: file: liboctave.a(rename.o) has no symbols
> ranlib: file: liboctave.a(rmdir.o) has no symbols
> ranlib: file: liboctave.a(strftime.o) has no symbols
> ranlib: file: liboctave.a(strptime.o) has no symbols
> ranlib: file: liboctave.a(strcasecmp.o) has no symbols
> ranlib: file: liboctave.a(strncase.o) has no symbols
> ranlib: file: liboctave.a(tempname.o) has no symbols
> ranlib: file: liboctave.a(tempnam.o) has no symbols
> ranlib: file: liboctave.a(oct-rl-edit.o) has no symbols
> ranlib: file: liboctave.a(oct-rl-hist.o) has no symbols
>
> feeling stumped...
>
>   
Is that really a problem.. Does the statically build octave run?

D.


From johny1888 at gmail.com  Sun Sep 13 05:59:30 2009
From: johny1888 at gmail.com (johny gaddar)
Date: Sun, 13 Sep 2009 16:29:30 +0530
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <mailman.1449.1252835583.21430.help-octave@octave.org>
References: <mailman.1449.1252835583.21430.help-octave@octave.org>
Message-ID: <f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>

Hello

i am new user to octave

i have a c program whose output which is two colums of data should be
plotted by octave as X and Y
i am using red hat machine.

any suggestions are appreciated.

regards
Johny
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/b2eb0717/attachment.html 

From lindnerben at gmx.net  Sun Sep 13 05:39:29 2009
From: lindnerben at gmx.net (Benjamin Lindner)
Date: Sun, 13 Sep 2009 12:39:29 +0200
Subject: imread error
In-Reply-To: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>
References: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>
Message-ID: <4AACCBE1.8050201@gmx.net>

Manish Kulkarni wrote:
> Dear everyone,
> 
> I'm getting the following error with imread
> 
>  > tftimage = imread("Waterfall.jpg");
> error: imread: invalid image file: Magick++ exception: Magick: Unable to 
> access configuration file (magic.mgk) reported
> by 
> d:/files/admin/octaveforge_svn/trunk/octave-forge/admin/Windows/mingw32/graphicsmagick/GraphicsMagick-1.3.3/magick/bl
> ob.c:1824 (GetConfigureBlob)
> error: called from:
> error:   C:\Octave\3.2.0_gcc-4.3.0\share\octave\3.2.0\m\image\imread.m 
> at line 74, column 7
> octave-3.2.0.exe:12:c:\Users\laptopGuru\Documents\mSilica\tftOctave
> 
> would anyone advise how to fix this?
> 

Update to the 3.2.2 binary. This issue has been fixed there.

benjamin


From rob.webinator at gmail.com  Sun Sep 13 07:04:35 2009
From: rob.webinator at gmail.com (Rob Redmon)
Date: Sun, 13 Sep 2009 08:04:35 -0400
Subject: Trouble installing Symbolic package, GiNaC, etc
Message-ID: <5514c9050909130504m307f8806sbccfc4e5fd5a085e@mail.gmail.com>

I'm getting these errors when trying to install the symbolic package:
octave-3.2.2:1> pkg install symbolic-1.0.9.tar.gz
symbols.cc:37:25: error: ginac/ginac.h: No such file or directory

My computer:
OS X, 10.5.8

I did this:
1) Downloaded and installed prebuilt Octave binary 3.2.2 for OS X from
http://octave.sourceforge.net

2) Realized I need the Symbolic library.

3) Installed GiNaC from MacPorts:
sudo port install ginac

--->  Fetching gmp
--->  Attempting to fetch gmp-4.2.4.tar.bz2 from
http://mirrors.ibiblio.org/pub/mirrors/gnu/ftp/gnu/gmp
--->  Verifying checksum(s) for gmp
--->  Extracting gmp
--->  Configuring gmp
--->  Building gmp
--->  Staging gmp into destroot
--->  Installing gmp @4.2.4_3
--->  Activating gmp @4.2.4_3
--->  Cleaning gmp
--->  Fetching cln
--->  Attempting to fetch cln-1.2.2.tar.bz2 from
http://distfiles.macports.org/cln
--->  Verifying checksum(s) for cln
--->  Extracting cln
--->  Configuring cln
--->  Building cln
--->  Staging cln into destroot
--->  Installing cln @1.2.2_0+test
--->  Activating cln @1.2.2_0+test
--->  Cleaning cln
--->  Fetching GiNaC
--->  Attempting to fetch ginac-1.4.4.tar.bz2 from
http://distfiles.macports.org/GiNaC
--->  Verifying checksum(s) for GiNaC
--->  Extracting GiNaC
--->  Configuring GiNaC
--->  Building GiNaC
--->  Staging GiNaC into destroot
--->  Installing GiNaC @1.4.4_0+test
--->  Activating GiNaC @1.4.4_0+test
--->  Cleaning GiNaC

4) Tried to install the symbolic library:
octave-3.2.2:1> pkg install symbolic-1.0.9.tar.gz
symbols.cc:37:25: error: ginac/ginac.h: No such file or directory
ov-vpa.h:39: error: ?GiNaC? has not been declared
ov-vpa.h:39: error: expected ?,? or ?...? before ?&? token
...
hundreds of errors follow

5) Do I need to build and install Octave from source?

Thanks,

Rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/3f9bd01f/attachment.html 

From dastew at sympatico.ca  Sun Sep 13 07:16:47 2009
From: dastew at sympatico.ca (dastew at sympatico.ca)
Date: Sun, 13 Sep 2009 12:16:47 +0000
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <f0693ce60909130429h151f2df7ye7d9c85028a10313@mail.gmail.com>
References: <mailman.1449.1252835583.21430.help-octave@octave.org>
	<f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>
	<COL107-W4149AE0C16E998590741D2A1E50@phx.gbl> 
	<f0693ce60909130429h151f2df7ye7d9c85028a10313@mail.gmail.com>
Message-ID: <COL107-W17E558F9D517D6C4982213A1E50@phx.gbl>


In octave do

pwd

(present working directory  pwd)

this is where octave will look for the file, so copy the data file to there and then octave will find it.

Doug



Date: Sun, 13 Sep 2009 16:59:33 +0530
Subject: Re: Help-octave Digest, Vol 42, Issue 28
From: johny1888 at gmail.com
To: dastew at sympatico.ca

thanks.

i am saving the output data of c program in data file named data.dat
now to load the file into octave, octave will need to know where the file is.
how i should tell octave where to find it
i saw the help on load, but it is not giving this info.


regards,
Johny

On Sun, Sep 13, 2009 at 4:39 PM,  <dastew at sympatico.ca> wrote:






One way to do it is to write the data to a text file, and then use the load command in octave to read the data from the file, and then plot it in octave.

Doug

Date: Sun, 13 Sep 2009 16:29:30 +0530
Subject: Re: Help-octave Digest, Vol 42, Issue 28

From: johny1888 at gmail.com
To: help-octave at octave.org


Hello

i am new user to octave

i have a c program whose output which is two colums of data should be plotted by octave as X and Y
i am using red hat machine.

any suggestions are appreciated.


regards

Johny



-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/eb085760/attachment.html 

From martin at mhelm.de  Sun Sep 13 07:44:01 2009
From: martin at mhelm.de (martin_helm)
Date: Sun, 13 Sep 2009 05:44:01 -0700 (PDT)
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <COL107-W17E558F9D517D6C4982213A1E50@phx.gbl>
References: <f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>
	<COL107-W17E558F9D517D6C4982213A1E50@phx.gbl>
Message-ID: <25422926.post@talk.nabble.com>




dastew wrote:
> 
> 
> In octave do
> 
> pwd
> 
> (present working directory  pwd)
> 
> this is where octave will look for the file, so copy the data file to
> there and then octave will find it.
> 
> Doug
> 
> 
> 
> Date: Sun, 13 Sep 2009 16:59:33 +0530
> Subject: Re: Help-octave Digest, Vol 42, Issue 28
> From: johny1888 at gmail.com
> To: dastew at sympatico.ca
> 
> thanks.
> 
> i am saving the output data of c program in data file named data.dat
> now to load the file into octave, octave will need to know where the file
> is.
> how i should tell octave where to find it
> i saw the help on load, but it is not giving this info.
> 
> 
> regards,
> Johny
> 
> On Sun, Sep 13, 2009 at 4:39 PM,  <dastew at sympatico.ca> wrote:
> 
> 
> 
> 
> 
> 
> One way to do it is to write the data to a text file, and then use the
> load command in octave to read the data from the file, and then plot it in
> octave.
> 
> Doug
> 
> Date: Sun, 13 Sep 2009 16:29:30 +0530
> Subject: Re: Help-octave Digest, Vol 42, Issue 28
> 
> From: johny1888 at gmail.com
> To: help-octave at octave.org
> 
> 
> Hello
> 
> i am new user to octave
> 
> i have a c program whose output which is two colums of data should be
> plotted by octave as X and Y
> i am using red hat machine.
> 
> any suggestions are appreciated.
> 
> 
> regards
> 
> Johny
> 
> 
> 
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

If you do not want to copy the file by hand you can load also tell to use a
path
for example

load -ascii "/path/to/datafile/data.dat"
plot(data(1,:), data(2,:))

- mh

-- 
View this message in context: http://www.nabble.com/Re%3A-Help-octave-Digest%2C-Vol-42%2C-Issue-28-tp25422282p25422926.html
Sent from the Octave - General mailing list archive at Nabble.com.


From lukas.reichlin at swissonline.ch  Sun Sep 13 09:32:42 2009
From: lukas.reichlin at swissonline.ch (Lukas Reichlin)
Date: Sun, 13 Sep 2009 16:32:42 +0200
Subject: OOP in Octave
Message-ID: <9830987E-56AA-4947-9FBA-C4C2700A0B1B@swissonline.ch>

Dear Octave Community

Since the control package needs to be rewritten in an oop fashion, I  
started to play around with some Matlab examples. However, Octave  
gives a parsing error when I try to run an example:

parse error near line 9 of file /Users/lukas/Desktop/polynom/ 
@DocPolynom/DocPolynom.m

   syntax error

 >>>    end
          ^

error: evaluating argument list element number 1
parse error near line 9 of file /Users/lukas/Desktop/polynom/ 
@DocPolynom/DocPolynom.m

   syntax error

 >>>    end
          ^

error: evaluating argument list element number 1
error: called from:
error:   /Users/lukas/Desktop/polynom/polytest.m at line 5, column 3
octave-3.2.2:5>

The code is from the MathWorks without any changes, it runs perfectly  
in Matlab, so it should be OK. What do I need to change to get it up  
and running in Octave?

Thanks for any help

Regards,
Lukas


classdef DocPolynom
    % Documentation example
    % A value class that implements a data type for polynonials
    % See Implementing a Class for Polynomials in the
    % MATLAB documentation for more information.

    properties
       coef
    end

    % Class methods
    methods
       function obj = DocPolynom(c)
          % Construct a DocPolynom object using the coefficients  
supplied
          if isa(c,'DocPolynom')
             obj.coef = c.coef;
          else
             obj.coef = c(:).';
          end
       end % DocPolynom
       function obj = set.coef(obj,val)
          ind = find(val(:).'~=0);
          if ~isempty(ind);
             obj.coef = val(ind(1):end);
          else
             obj.coef = val;
          end
       end

       function c = double(obj)
          c = obj.coef;
       end % double

       function s = char(obj)
          % Created a formated display of the polynom
          % as powers of x
          if all(obj.coef == 0)
             s = '0';
          else
             d = length(obj.coef)-1;
             s = cell(1,d);
             ind = 1;
             for a = obj.coef;
                if a ~= 0;
                   if ind ~= 1
                      if a > 0
                         s(ind) = {' + '};
                         ind = ind + 1;
                      else
                         s(ind) = {' - '};
                         a = -a;
                         ind = ind + 1;
                      end
                   end
                   if a ~= 1 || d == 0
                      if a == -1
                         s(ind) = {'-'};
                         ind = ind + 1;
                      else
                         s(ind) = {num2str(a)};
                         ind = ind + 1;
                         if d > 0
                            s(ind) = {'*'};
                            ind = ind + 1;
                         end
                      end
                   end
                   if d >= 2
                      s(ind) = {['x^' int2str(d)]};
                      ind = ind + 1;
                   elseif d == 1
                      s(ind) = {'x'};
                      ind = ind + 1;
                   end
                end
                d = d - 1;
             end
          end
       end % char

       function disp(obj)
          % DISP Display object in MATLAB syntax
          c = char(obj);
          if iscell(c)
             disp(['     ' c{:}])
          else
             disp(c)
          end
       end % disp

       function b = subsref(a,s)
          % SUBSREF Implementing the following syntax:
          % obj([1 ...])
          % obj.coef
          % obj.plot
          % out = obj.method(args)
          % out = obj.method
          switch s(1).type
             case '()'
                ind = s.subs{:};
                b = a.polyval(ind);
             case '.'
                switch s(1).subs
                   case 'coef'
                      b = a.coef;
                   case 'plot'
                      a.plot;
                   otherwise
                      if length(s)>1
                         b = a.(s(1).subs)(s(2).subs{:});
                      else
                         b = a.(s.subs);
                      end
                end
             otherwise
                error('Specify value for x as obj(x)')
          end
       end % subsref

       function r = plus(obj1,obj2)
          % PLUS  Implement obj1 + obj2 for DocPolynom
          obj1 = DocPolynom(obj1);
          obj2 = DocPolynom(obj2);
          k = length(obj2.coef) - length(obj1.coef);
          r = DocPolynom([zeros(1,k) obj1.coef] + [zeros(1,-k)  
obj2.coef]);
       end % plus

       function r = minus(obj1,obj2)
          % MINUS Implement obj1 - obj2 for DocPolynoms.
          obj1 = DocPolynom(obj1);
          obj2 = DocPolynom(obj2);
          k = length(obj2.coef) - length(obj1.coef);
          r = DocPolynom([zeros(1,k) obj1.coef] - [zeros(1,-k)  
obj2.coef]);
       end % minus

       function r = mtimes(obj1,obj2)
          % MTIMES   Implement obj1 * obj2 for DocPolynoms.
          obj1 = DocPolynom(obj1);
          obj2 = DocPolynom(obj2);
          r = DocPolynom(conv(obj1.coef,obj2.coef));
       end % mtimes

       function r = roots(obj)
          % ROOTS.  ROOTS(obj) is a vector containing the roots of obj.
          r = roots(obj.coef);
       end % roots

       function y = polyval(obj,x)
          % POLYVAL  POLYVAL(obj,x) evaluates obj at the points x.
          y = polyval(obj.coef,x);
       end % polyval

       function q = diff(obj)
          % DIFF  DIFF(obj) is the derivative of the polynom obj.
          c = obj.coef;
          d = length(c) - 1;  % degree
          q = DocPolynom(obj.coef(1:d).*(d:-1:1));
       end % diff

       function plot(obj)
          % PLOT  PLOT(obj) plots the polynom obj
          r = max(abs(roots(obj)));
          x = (-1.1:0.01:1.1)*r;
          y = polyval(obj,x);
          plot(x,y);
          c = char(obj);
          title(['y = ' c{:}])
          xlabel('X')
          ylabel('Y','Rotation',0)
          grid on
       end % plot
    end % methods
end % classdef

-------------- next part --------------
A non-text attachment was scrubbed...
Name: polynom.zip
Type: application/zip
Size: 1886 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/b5019f92/attachment-0001.zip 

From johny1888 at gmail.com  Sun Sep 13 10:13:47 2009
From: johny1888 at gmail.com (johny gaddar)
Date: Sun, 13 Sep 2009 20:43:47 +0530
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <200909131354.32995.martin@mhelm.de>
References: <mailman.1449.1252835583.21430.help-octave@octave.org>
	<f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>
	<200909131354.32995.martin@mhelm.de>
Message-ID: <f0693ce60909130813o7aeddb1fjd03818e147bd8d73@mail.gmail.com>

yes, i am storing the output of the program in .dat kind of file

the file contents look like this.....


0 1
1 3
4 7
7 8

and so on.
there are only two columns. of integers. thats all.
i want to plot second column against first column in octave.
other than this, file does not contain anything else.

regards,
Johny


On Sun, Sep 13, 2009 at 5:24 PM, Martin Helm <martin at mhelm.de> wrote:

> Am Sonntag, 13. September 2009 12:59:30 schrieb johny gaddar:
> > Hello
> >
> > i am new user to octave
> >
> > i have a c program whose output which is two colums of data should be
> > plotted by octave as X and Y
> > i am using red hat machine.
> >
> > any suggestions are appreciated.
> >
> > regards
> > Johny
>
> Dear Johny,
>
> from your description I asume that your program creates some file which you
> want to read in octave to plot it (is that correct).
> Can you provide a simple example of such file to see the format of the
> numbers
> how they are delimited and if the file contains some header?
> Then I can give you a hint how to read and display it.
>
> In general you can for example look at csvread and similar functions
>
> help csvread
> `csvread' is a function from the file
> /usr/share/octave/3.2.0/m/io/csvread.m
>
>  -- Function File: X = csvread (FILENAME)
>     Read the matrix X from a file.
>
>     This function is equivalent to
>          dlmread (FILENAME, "," , ...)
>
>     See also: dlmread, dlmwrite, csvwrite
>
> - mh
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/a9478ba2/attachment.html 

From martin at mhelm.de  Sun Sep 13 10:38:56 2009
From: martin at mhelm.de (Martin Helm)
Date: Sun, 13 Sep 2009 17:38:56 +0200
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <f0693ce60909130813o7aeddb1fjd03818e147bd8d73@mail.gmail.com>
References: <mailman.1449.1252835583.21430.help-octave@octave.org>
	<200909131354.32995.martin@mhelm.de>
	<f0693ce60909130813o7aeddb1fjd03818e147bd8d73@mail.gmail.com>
Message-ID: <200909131738.56949.martin@mhelm.de>

Am Sonntag, 13. September 2009 17:13:47 schrieb johny gaddar:
> yes, i am storing the output of the program in .dat kind of file
>
> the file contents look like this.....
>
>
> 0 1
> 1 3
> 4 7
> 7 8
>
> and so on.
> there are only two columns. of integers. thats all.
> i want to plot second column against first column in octave.
> other than this, file does not contain anything else.
>
> regards,
> Johny
>

So the form

load -ascii "/path/to/datafile/data.dat" 
plot(data(1,:), data(2,:)) 

should work (assuming data.dat) otherwise the resulting variable has simply a 
different name

mymatrix.dat -> mymatrix cointains the data and so on. 


> On Sun, Sep 13, 2009 at 5:24 PM, Martin Helm <martin at mhelm.de> wrote:
> > Am Sonntag, 13. September 2009 12:59:30 schrieb johny gaddar:
> > > Hello
> > >
> > > i am new user to octave
> > >
> > > i have a c program whose output which is two colums of data should be
> > > plotted by octave as X and Y
> > > i am using red hat machine.
> > >
> > > any suggestions are appreciated.
> > >
> > > regards
> > > Johny
> >
> > Dear Johny,
> >
> > from your description I asume that your program creates some file which
> > you want to read in octave to plot it (is that correct).
> > Can you provide a simple example of such file to see the format of the
> > numbers
> > how they are delimited and if the file contains some header?
> > Then I can give you a hint how to read and display it.
> >
> > In general you can for example look at csvread and similar functions
> >
> > help csvread
> > `csvread' is a function from the file
> > /usr/share/octave/3.2.0/m/io/csvread.m
> >
> >  -- Function File: X = csvread (FILENAME)
> >     Read the matrix X from a file.
> >
> >     This function is equivalent to
> >          dlmread (FILENAME, "," , ...)
> >
> >     See also: dlmread, dlmwrite, csvwrite
> >
> > - mh


From martin at mhelm.de  Sun Sep 13 06:54:32 2009
From: martin at mhelm.de (Martin Helm)
Date: Sun, 13 Sep 2009 13:54:32 +0200
Subject: Help-octave Digest, Vol 42, Issue 28
In-Reply-To: <f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>
References: <mailman.1449.1252835583.21430.help-octave@octave.org>
	<f0693ce60909130359w4fe386adm2d1b025334050ce0@mail.gmail.com>
Message-ID: <200909131354.32995.martin@mhelm.de>

Am Sonntag, 13. September 2009 12:59:30 schrieb johny gaddar:
> Hello
>
> i am new user to octave
>
> i have a c program whose output which is two colums of data should be
> plotted by octave as X and Y
> i am using red hat machine.
>
> any suggestions are appreciated.
>
> regards
> Johny

Dear Johny,

from your description I asume that your program creates some file which you 
want to read in octave to plot it (is that correct).
Can you provide a simple example of such file to see the format of the numbers 
how they are delimited and if the file contains some header?
Then I can give you a hint how to read and display it.

In general you can for example look at csvread and similar functions 

help csvread
`csvread' is a function from the file /usr/share/octave/3.2.0/m/io/csvread.m

 -- Function File: X = csvread (FILENAME)
     Read the matrix X from a file.

     This function is equivalent to
          dlmread (FILENAME, "," , ...)

     See also: dlmread, dlmwrite, csvwrite

- mh

From dbateman at dbateman.org  Sun Sep 13 12:08:09 2009
From: dbateman at dbateman.org (David Bateman)
Date: Sun, 13 Sep 2009 19:08:09 +0200
Subject: OOP in Octave
In-Reply-To: <9830987E-56AA-4947-9FBA-C4C2700A0B1B@swissonline.ch>
References: <9830987E-56AA-4947-9FBA-C4C2700A0B1B@swissonline.ch>
Message-ID: <4AAD26F9.100@dbateman.org>

Lukas Reichlin wrote:
> Dear Octave Community
>
> Since the control package needs to be rewritten in an oop fashion, I 
> started to play around with some Matlab examples. However, Octave 
> gives a parsing error when I try to run an example:
>
> parse error near line 9 of file 
> /Users/lukas/Desktop/polynom/@DocPolynom/DocPolynom.m
>
>   syntax error
>
> >>>    end
>          ^
>
> error: evaluating argument list element number 1
> parse error near line 9 of file 
> /Users/lukas/Desktop/polynom/@DocPolynom/DocPolynom.m
>
>   syntax error
>
> >>>    end
>          ^
>
> error: evaluating argument list element number 1
> error: called from:
> error:   /Users/lukas/Desktop/polynom/polytest.m at line 5, column 3
> octave-3.2.2:5>
>
> The code is from the MathWorks without any changes, it runs perfectly 
> in Matlab, so it should be OK. What do I need to change to get it up 
> and running in Octave?
>
> Thanks for any help
>
> Regards,
> Lukas
>
>
> classdef DocPolynom
There is no classdef in Octave yet, though some preliminary work is 
done.. There is only the old matlab OOP style code in octave (ie tough 
with the @ directories). The matlab docs for this style of coding are at

http://www.mathworks.com/access/helpdesk/help/pdf_doc/matlab/pre-version_7.6_oop.pdf

and there is a chapter in the Octave manual in 3.2.2 on the OOP support 
in octave

D.


-- 
David Bateman                                dbateman at dbateman.org
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)


From phil at beadling.co.uk  Sun Sep 13 16:19:11 2009
From: phil at beadling.co.uk (Philip Beadling)
Date: Sun, 13 Sep 2009 22:19:11 +0100
Subject: Yapso scales surf plots wrong?
Message-ID: <1252876751.7374.26.camel@phil-desktop>

Hi all,

I Stumbled across Yapso today - it has some very nice plotting
functionality.  I can get the examples working fine, but when I try to
plot some of my own data I get a scaling issue.

My data ranges are created using linspace like:

n=50
x = linspace(0,2,n)
y = linspace(20,80,n)
[X,Y] = meshgrid(x,y)
% Z's range is approx 30-50.
Z = func(X,Y)
surf(X,Y,Z,Z)
shading faceted
colorbar

Using the standard Gnuplot this is scaled perfectly.  But with Yapso the
same scale is used for all 3 axis (as far as I can tell).  Hence the x
axis being from 0 to 2 is compressed to the point where the graph looks
really messed up.

First question is - why doesn't Yapso scale automatically, am I missing
something?  I note all the examples on the webpage seem to only have
examples where scales are on the same order of magnitude, so there are
no examples to confirm my issue.

Secondly, if I need to specify that x axis has a much finer granularity
scale how do I do that?


Thanks very much,

Phil



From manishdkulkarni at gmail.com  Sun Sep 13 23:39:15 2009
From: manishdkulkarni at gmail.com (Manish Kulkarni)
Date: Sun, 13 Sep 2009 21:39:15 -0700
Subject: imread error
In-Reply-To: <4AACCBE1.8050201@gmx.net>
References: <8feb35460909111257nc1a42bx7868a4e430d98481@mail.gmail.com>
	<4AACCBE1.8050201@gmx.net>
Message-ID: <8feb35460909132139w2d87af10r1fcfa177910c6779@mail.gmail.com>

Thanks a lot Benjamin, will try this out.

Cheers
Manish

On Sun, Sep 13, 2009 at 3:39 AM, Benjamin Lindner <lindnerben at gmx.net>wrote:

> Manish Kulkarni wrote:
>
>> Dear everyone,
>>
>> I'm getting the following error with imread
>>
>>  > tftimage = imread("Waterfall.jpg");
>> error: imread: invalid image file: Magick++ exception: Magick: Unable to
>> access configuration file (magic.mgk) reported
>> by
>> d:/files/admin/octaveforge_svn/trunk/octave-forge/admin/Windows/mingw32/graphicsmagick/GraphicsMagick-1.3.3/magick/bl
>> ob.c:1824 (GetConfigureBlob)
>> error: called from:
>> error:   C:\Octave\3.2.0_gcc-4.3.0\share\octave\3.2.0\m\image\imread.m at
>> line 74, column 7
>> octave-3.2.0.exe:12:c:\Users\laptopGuru\Documents\mSilica\tftOctave
>>
>> would anyone advise how to fix this?
>>
>>
> Update to the 3.2.2 binary. This issue has been fixed there.
>
> benjamin
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090913/d7193eb8/attachment.html 

From linuxdieg at gmail.com  Mon Sep 14 01:43:35 2009
From: linuxdieg at gmail.com (diego vega)
Date: Mon, 14 Sep 2009 01:43:35 -0500
Subject: help me
Message-ID: <fbe9db0c0909132343w10556861t51a5859271c59061@mail.gmail.com>

comunidad de octave, cordial saludo
me dirijo a ustedes poque tengo una duda acerca de ciertas funciones que no
he podido usar en octave, quisiera saber si existen o hay que crearlas;
las funciones son: funcion escalon, funcion signo, pulso triangular ,
inpulso unitario.

muchas gracias
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090914/52caad67/attachment.html 

From lukas.reichlin at swissonline.ch  Mon Sep 14 02:01:33 2009
From: lukas.reichlin at swissonline.ch (Lukas Reichlin)
Date: Mon, 14 Sep 2009 09:01:33 +0200
Subject: OOP in Octave
In-Reply-To: <4AAD26F9.100@dbateman.org>
References: <9830987E-56AA-4947-9FBA-C4C2700A0B1B@swissonline.ch>
	<4AAD26F9.100@dbateman.org>
Message-ID: <7683DFC4-C36F-408E-A4E7-1A03AD77E352@swissonline.ch>


On 13.09.2009, at 19:08, David Bateman wrote:

> Lukas Reichlin wrote:
>> Dear Octave Community
>>
>> Since the control package needs to be rewritten in an oop fashion,  
>> I started to play around with some Matlab examples. However, Octave  
>> gives a parsing error when I try to run an example:
>>
>> parse error near line 9 of file /Users/lukas/Desktop/polynom/ 
>> @DocPolynom/DocPolynom.m
>>
>>  syntax error
>>
>> >>>    end
>>         ^
>>
>> error: evaluating argument list element number 1
>> parse error near line 9 of file /Users/lukas/Desktop/polynom/ 
>> @DocPolynom/DocPolynom.m
>>
>>  syntax error
>>
>> >>>    end
>>         ^
>>
>> error: evaluating argument list element number 1
>> error: called from:
>> error:   /Users/lukas/Desktop/polynom/polytest.m at line 5, column 3
>> octave-3.2.2:5>
>>
>> The code is from the MathWorks without any changes, it runs  
>> perfectly in Matlab, so it should be OK. What do I need to change  
>> to get it up and running in Octave?
>>
>> Thanks for any help
>>
>> Regards,
>> Lukas
>>
>>
>> classdef DocPolynom
> There is no classdef in Octave yet, though some preliminary work is  
> done.. There is only the old matlab OOP style code in octave (ie  
> tough with the @ directories). The matlab docs for this style of  
> coding are at
>
> http://www.mathworks.com/access/helpdesk/help/pdf_doc/matlab/pre-version_7.6_oop.pdf
>
> and there is a chapter in the Octave manual in 3.2.2 on the OOP  
> support in octave
>
> D.
>
>
> -- 
> David Bateman                                dbateman at dbateman.org
> 35 rue Gambetta                              +33 1 46 04 02 18 (Home)
> 92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)
>

Thanks for your quick reply. Do you know when the new style should be  
ready to use? (in one year, in two years or Octave 3.4, 3.6 ...)

Regards
Lukas


From highegg at gmail.com  Mon Sep 14 02:31:08 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 14 Sep 2009 09:31:08 +0200
Subject: Wrapper fortran
In-Reply-To: <25408339.post@talk.nabble.com>
References: <25408339.post@talk.nabble.com>
Message-ID: <69d8d540909140031uee193cj2a8f0d00bf40763b@mail.gmail.com>

On Fri, Sep 11, 2009 at 11:07 PM, R.C. <romain.cotte at usherbrooke.ca> wrote:
>
> Hello,
>
>
> I followed example ?from http://wiki.octave.org/wiki.pl?OctaveFortran but I
> have some trouble with my fortran subroutine.
> Compilation works but when i call my function it gives ;
>
> [f, xb] = wrapper (2, 4, rand(4,1), 1)
> *** glibc detected *** octave: free(): invalid next size (fast): 0x09c20450
> ***
> ======= Backtrace: =========
> /lib/tls/i686/cmov/libc.so.6[0xb5d31604]
> /lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb5d335b6]
> /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0xb5eed231]
> /usr/lib/libstdc++.so.6(_ZdaPv+0x1d)[0xb5eed28d]
> /usr/lib/octave-3.0.1/liboctinterp.so(_ZN5ArrayIdED2Ev+0xc6)[0xb7a30c66]
> etc...
>
>
> My wrapper
> ***********************************************************************************
> #include <octave/oct.h>
> #include "f77-fcn.h"
> extern "C"
> ?{
> ? ?int F77_FUNC (rosenf_b, ROSENF_b) (const int& ind, const int& n,
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?const double* x,
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?double* xb, double& f,
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?const double& fb,
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?double& g, double& ti, double& tr, double& td );
> ?}
> ?DEFUN_DLD (wrapper, args, ,
> ? ? ? ? ? ?"- Loadable Function: [f, xb] = wrapper (ind, n, x, fb)\n\
> ?\n\
> ?Returns the f,xb.")
> ?{
> ? ?octave_value_list retval;
> ? ?Matrix X;
>
> ? ?const int IND = args(0).int_value(), N = args(1).int_value();
> ? ?const double FB;
> ? ?double G;
> ? ?NDArray XB;
> ? ?double F,TI,TR,TD;
> ? ?double *av;
>
> ? ? FB = args(3).double_value();
> ? ? X = args(2).matrix_value();
> ? ? av = XB.fortran_vec();
>
>
> ? ?F77_XFCN (rosenf_b, ROSENF_B,
> ? ? ? ? ? ?(IND, N,
> ? ? ? ? ? ?X.fortran_vec(),
> ? ? ? ? ? ?av,
> ? ? ? ? ? ?F,
> ? ? ? ? ? ?FB,
> ? ? ? ? ? ?G,
> ? ? ? ? ? ?TI, TR, TD) );
> ? ?if (f77_exception_encountered)
> ? ? ?{
> ? ? ? ?error ("unrecoverable error in rosenf_b");
> ? ? ? ?return retval;
> ? ? ?}
> ? ?retval(0) = octave_value (F);
> ? ?retval(1) = octave_value(XB);
>
> ? ?return retval;
> ?}
> *************************************************************************
>
>
>
>
> Fortran function
> ************************************************************************
> ? ? ?SUBROUTINE ROSENF_B(ind, n, x, xb, f, fb, g, ti, tr, td)
> ? ? ?IMPLICIT NONE
> ? ? ?INTEGER ind, n, ti(*)
> ? ? ?DOUBLE PRECISION x(n), f, g(n), td(*)
> ? ? ?DOUBLE PRECISION xb(n), fb
> ? ? ?REAL tr(*)
> C
> ? ? ?DOUBLE PRECISION y, p
> ? ? ?INTEGER i
> ? ? ?DOUBLE PRECISION tempb
> ? ? ?INTEGER ii1
> ? ? ?p = 100.0d0
> ? ? ?IF (ind .EQ. 2 .OR. ind .EQ. 4) THEN
> ? ? ? ?DO ii1=1,n
> ? ? ? ? ?xb(ii1) = 0.D0
> ? ? ? ?ENDDO
> ? ? ? ?DO i=n,2,-1
> ? ? ? ? ?tempb = p*2*(x(i)-x(i-1)**2)*fb
> ? ? ? ? ?xb(i) = xb(i) + tempb - 2*(1.0d0-x(i))*fb
> ? ? ? ? ?xb(i-1) = xb(i-1) - 2*x(i-1)*tempb
> ? ? ? ?ENDDO
> ? ? ? ?fb = 0.D0
> ? ? ?ELSE
> ? ? ? ?DO ii1=1,n
> ? ? ? ? ?xb(ii1) = 0.D0
> ? ? ? ?ENDDO
> ? ? ?END IF
> ? ? ?END
> **********************************************************************
>
>
>
> What do you suggest ?
> Thanks
>

To fix your code. XB needs to be allocated (av points to buffer of
size 0 in your example).
Btw., it makes little sense to pass n to the function, when it can be
inferred from x.

hth

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From highegg at gmail.com  Mon Sep 14 04:41:19 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 14 Sep 2009 11:41:19 +0200
Subject: Using STR2FUNC in sub-function to refer to other sub-function(s)
In-Reply-To: <47AC1666471FFC41873637D50D111B812679568394@SINTEFEXMBX01.sintef.no>
References: <AQHKMuepngfEfgXod0iQC+5Mm9w96g==>
	<47AC1666471FFC41873637D50D111B812679568394@SINTEFEXMBX01.sintef.no>
Message-ID: <69d8d540909140241m4f52be6ek847b8c574f993e8c@mail.gmail.com>

On Fri, Sep 11, 2009 at 4:04 PM, B?rd Skaflestad
<Bard.Skaflestad at sintef.no> wrote:
> All,
>
> I'm experiencing a bit of bother using the STR2FUNC function. ?For sake of argument I'll use the synthetic function 'foo' defined as
>
> $ cat foo.m
> ? function h = foo()
> ? ? ?h = @sub1;
> ? end
>
> ? function h = sub1()
> ? ? ?h = str2func('sub2');
> ? end
>
> ? function h = sub2()
> ? ? ?h = @(x) sub3(x, 10);
> ? end
>
> ? function q = sub3(a, b)
> ? ? ?q = fprintf('sub3: (a,b) = (%f,%d)\n', a, b);
> ? end
> $
>
> In Octave I then get
>
> --------------------8<----------------------------------------8<-------------------------
>
> ? ?octave:29> f = foo(), s1 = f(), s2 = s1(), s3 = s2(pi)
> ? ?f =
>
> ? ?sub1
>
> ? ?error: error creating function handle "@sub2"
> ? ?error: called from:
> ? ?error: ? /home/bska/tmp/build/octave/3.2.2/foo.m at line 6, column 9
>
> --------------------8<----------------------------------------8<-------------------------
>
> while MATLAB yields the, to me, expected output
>
> --------------------8<----------------------------------------8<-------------------------
>
> ? ?>> f = foo(), s1 = f(), s2 = s1(), s3 = s2(pi)
>
> ? ?f =
>
> ? ? ? ?@sub1
>
> ? ?s1 =
>
> ? ? ? ?@sub2
>
> ? ?s2 =
>
> ? ? ? ?@(x)sub3(x,10)
>
> ? ?sub3: (a,b) = (3.141593,10)
>
> ? ?s3 =
>
> ? ? ? ?28
>
> --------------------8<----------------------------------------8<-------------------------
>
> My question is if this (failing to construct a function handle to 'sub2') is an error in Octave or if Octave's STR2FUNC implementation is not expected to have the ability to construct handles to (other) sub-functions of a "main" function. ?I should note that if I replace the above 'sub1' function by the (equivalent, in this case)
>
> ? ?function h = sub1()
> ? ? ? h = @sub2;
> ? ?end
>
> the above error goes away. ?In the general case, however, I would really like the ability to use STR2FUNC here because the function name is not generally known until runtime in code. ?Admittedly, my code uses only a bounded set of possible string values (roughly 5-10 alternatives), but I'd rather not insert a big SWITCH statement in there. ?Any and all suggestions will be greatly appreciated.
>
>
> Best regards,
> --
> B?rd Skaflestad ? ? ? ? ? ? ? ? ? ? ? ? ? ?<Bard.Skaflestad at sintef.no>
> SINTEF ICT, Applied mathematics
> Forskningsveien 1, No-0314 Oslo
> Tel.: (+47) 2206 7642
>
>

The following patch fixes the problem in the development tree:
http://hg.savannah.gnu.org/hgweb/octave/rev/8d79f36ebdde

this patch is not directly applicable to 3.2.x (breaks ABI), so it
will need a separate patch. It may not make it into 3.2.3 which is
just days away.

thanks

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From Bard.Skaflestad at sintef.no  Mon Sep 14 06:53:38 2009
From: Bard.Skaflestad at sintef.no (=?ISO-8859-1?Q?B=E5rd?= Skaflestad)
Date: Mon, 14 Sep 2009 13:53:38 +0200
Subject: Using STR2FUNC in sub-function to refer to other sub-function(s)
In-Reply-To: <69d8d540909140241m4f52be6ek847b8c574f993e8c@mail.gmail.com>
References: <AQHKMuepngfEfgXod0iQC+5Mm9w96g==>
	<47AC1666471FFC41873637D50D111B812679568394@SINTEFEXMBX01.sintef.no>
	<69d8d540909140241m4f52be6ek847b8c574f993e8c@mail.gmail.com>
Message-ID: <1252929218.14408.1.camel@skuffe.math.sintef.no>

On Mon, 2009-09-14 at 11:41 +0200, Jaroslav Hajek wrote:
> On Fri, Sep 11, 2009 at 4:04 PM, B?rd Skaflestad
> <Bard.Skaflestad at sintef.no> wrote:

[A problem description]

> The following patch fixes the problem in the development tree:
> http://hg.savannah.gnu.org/hgweb/octave/rev/8d79f36ebdde
> 
> this patch is not directly applicable to 3.2.x (breaks ABI), so it
> will need a separate patch. It may not make it into 3.2.3 which is
> just days away.
> 
> thanks

Thank you so much for the swift response and solution.  I'll try the
development tree later this week.


Best regards,
-- 
B?rd Skaflestad                            <Bard.Skaflestad at sintef.no>
SINTEF ICT, Applied mathematics
Forskningsveien 1, No-0314 Oslo
Tel.: (+47) 2206 7642



From christy1865 at yahoo.com  Mon Sep 14 07:06:25 2009
From: christy1865 at yahoo.com (Kristen Richter)
Date: Mon, 14 Sep 2009 05:06:25 -0700 (PDT)
Subject: octave function 'avifile' help
Message-ID: <111889.53914.qm@web30208.mail.mud.yahoo.com>

Hello all,

I am attempting to create a movie from a series of plots using octave function 'avifile'.  I found this function in the video add-on package
available from Octave-forge, but I am having some trouble using it because octave  can't seem to find the appropriate codecs.  I am trying out some example code I found in the mailing list archives.
Link:http://neacm.fe.up.pt/pub/octave/MAILING-LISTS/help-octave/2005/3694

I have pasted the code below:

# Open the AVI file
m = avifile("test.avi", "codec", "msmpeg4v2")

# Generate and add frames
for i = 1:100
  I = zeros(100,100);
  I(i,:) = i;
  I(:,i) = 200-i;
  addframe(m, I/255)
  printf(".")
endfor
printf("\n")

# Close the file
clear m

However, after running these commands on the octave command line, I get the following error:

AVHandler: cannot find codec msmpeg4v2
error: avifile: AVHandler setup failed
addframe: error adding frame to avifile

I've
tried the example, specifying different kinds of audio codecs, including cinepak,
indeo5, and indeo3.  I've checked my synaptic package manager to make sure I have these audio  codecs installed.  During troubleshooting, I've installed libavifile (vs. 0.7),
which mentions the indeo library by name.

Are the codecs supposed to be installed in a specific directory? Are they supposed to be referred to by a text string that also contains the directory name? Has anyone come across similar issues in the past?

The example is pretty old... dated 2005, and I know the video package was updated last this past May.

Any insight on the situation would be greatly appreciated. As always, your help in the past has been invaluable.

-Kristen Richter



      

From mangostein17 at googlemail.com  Mon Sep 14 10:33:03 2009
From: mangostein17 at googlemail.com (matt)
Date: Mon, 14 Sep 2009 16:33:03 +0100
Subject: octave vs. matlab vectorization
Message-ID: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>

i have used octave exclusively for a number of years and just  
recently had to use matlab for teaching purposes.

since i was raised to vectorize loop structures whenever possible, i  
thought i would illustrate the the speed-up for the students in the  
class by comparing execution time for two codes--code 1 with loops,  
and code 2 vectorized.

here is code 1:

t0=clock;
for m=1:1000000
	q=zeros(1,10);
	for k=1:10,
     		tmp=k^2;
     		if tmp>50,
         		q(k)=tmp;
     		else
         		q(k)=0;
     		end
	end
end
etime(clock,t0)

here is code 2:

t0=clock;
for m=1:1000000
	q=zeros(1,10);
	tmp=(1:10).^2;
	q=tmp.*(tmp>50);
end
etime(clock,t0)

on my laptop, octave (3.0.5) takes 200 secs for the first code and  
just 24 secs for the second--i thought i had proved my point.

however, on the department server (admittedly a much faster machine),  
matlab ground through the first code in 0.82 secs while the second  
code took 4.3 secs.

so, my question: has matlab been optimized for loops? is there some  
sort of compilation going on? any insight gratefully received. i have  
been writing code in octave for a while now and had assumed it would  
almost be a straight port to matlab.

cheers,

matt.


From storrsjm at email.uc.edu  Mon Sep 14 11:07:33 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Mon, 14 Sep 2009 12:07:33 -0400
Subject: octave vs. matlab vectorization
In-Reply-To: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
Message-ID: <a0daad600909140907w6bf73f13ie899cf58d0d22c71@mail.gmail.com>

Matlab has used a JIT compiler for a while now so the vectorization penalty
in loops often isn't nearly as severe as it once was. In the Q&A of a sales
presentation a few years ago, I remember it was mentioned that Matlab's JIT
compiler is able to recognize and vectorize some loops more efficiently than
others. Octave doesn't have JIT.

--judd


On Mon, Sep 14, 2009 at 11:33 AM, matt <mangostein17 at googlemail.com> wrote:

> i have used octave exclusively for a number of years and just
> recently had to use matlab for teaching purposes.
>
> since i was raised to vectorize loop structures whenever possible, i
> thought i would illustrate the the speed-up for the students in the
> class by comparing execution time for two codes--code 1 with loops,
> and code 2 vectorized.
>
> here is code 1:
>
> t0=clock;
> for m=1:1000000
>        q=zeros(1,10);
>        for k=1:10,
>                tmp=k^2;
>                if tmp>50,
>                        q(k)=tmp;
>                else
>                        q(k)=0;
>                end
>        end
> end
> etime(clock,t0)
>
> here is code 2:
>
> t0=clock;
> for m=1:1000000
>        q=zeros(1,10);
>        tmp=(1:10).^2;
>        q=tmp.*(tmp>50);
> end
> etime(clock,t0)
>
> on my laptop, octave (3.0.5) takes 200 secs for the first code and
> just 24 secs for the second--i thought i had proved my point.
>
> however, on the department server (admittedly a much faster machine),
> matlab ground through the first code in 0.82 secs while the second
> code took 4.3 secs.
>
> so, my question: has matlab been optimized for loops? is there some
> sort of compilation going on? any insight gratefully received. i have
> been writing code in octave for a while now and had assumed it would
> almost be a straight port to matlab.
>
> cheers,
>
> matt.
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090914/4823a2b5/attachment.html 

From Potorti at isti.cnr.it  Mon Sep 14 11:31:35 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Mon, 14 Sep 2009 18:31:35 +0200
Subject: octave vs. matlab vectorization
In-Reply-To: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
Message-ID: <E1MnESl-0002Bd-1r@tucano.isti.cnr.it>

>since i was raised to vectorize loop structures whenever possible, i  
>thought i would illustrate the the speed-up for the students in the  
>class by comparing execution time for two codes--code 1 with loops,  
>and code 2 vectorized.
>
>here is code 1:
>
>t0=clock;
>for m=1:1000000
>	q=zeros(1,10);
>	for k=1:10,
>     		tmp=k^2;
>     		if tmp>50,
>         		q(k)=tmp;
>     		else
>         		q(k)=0;
>     		end
>	end
>end
>etime(clock,t0)
>
>here is code 2:
>
>t0=clock;
>for m=1:1000000
>	q=zeros(1,10);
>	tmp=(1:10).^2;
>	q=tmp.*(tmp>50);
>end
>etime(clock,t0)
>
>on my laptop, octave (3.0.5) takes 200 secs for the first code and  
>just 24 secs for the second--i thought i had proved my point.

A ratio of 8.3:1

>however, on the department server (admittedly a much faster machine),  
>matlab ground through the first code in 0.82 secs while the second  
>code took 4.3 secs.

A ratio of 5.2:1.  Not quite the same, but I think the point is proven
nonetheless.

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From mangostein17 at googlemail.com  Mon Sep 14 11:57:10 2009
From: mangostein17 at googlemail.com (matt)
Date: Mon, 14 Sep 2009 17:57:10 +0100
Subject: octave vs. matlab vectorization
In-Reply-To: <E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
Message-ID: <BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>

ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1

i find this result quite staggering.

On 14 Sep 2009, at 17:31, Francesco Potorti` wrote:

>>
>> on my laptop, octave (3.0.5) takes 200 secs for the first code and
>> just 24 secs for the second--i thought i had proved my point.
>
> A ratio of 8.3:1
>
>> however, on the department server (admittedly a much faster machine),
>> matlab ground through the first code in 0.82 secs while the second
>> code took 4.3 secs.
>
> A ratio of 5.2:1.  Not quite the same, but I think the point is proven
> nonetheless.
>
> --  
> Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op. 
> 2111)
> ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
> via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
> (entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/



From jwe at octave.org  Mon Sep 14 12:29:52 2009
From: jwe at octave.org (John W. Eaton)
Date: Mon, 14 Sep 2009 13:29:52 -0400
Subject: octave vs. matlab vectorization
In-Reply-To: <BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
Message-ID: <19118.32144.109057.627543@segfault.lan>

On 14-Sep-2009, matt wrote:

| ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
| 
| i find this result quite staggering.

Why?  As someone else already explained, Matlab has a JIT compiler
that can, in some cases at least, compile loops into native code so
that it runs faster and Octave does not have a JIT compiler.

jwe

From shermanj at umd.edu  Mon Sep 14 12:43:54 2009
From: shermanj at umd.edu (James Sherman Jr.)
Date: Mon, 14 Sep 2009 13:43:54 -0400
Subject: octave vs. matlab vectorization
In-Reply-To: <19118.32144.109057.627543@segfault.lan>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<19118.32144.109057.627543@segfault.lan>
Message-ID: <6800c3890909141043g32e6891ds4bde74e8bc78ec71@mail.gmail.com>

>ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
>
>i find this result quite staggering.

I ran those two codes with the matlab I have on my laptop (7.7.0.471 R2008b)
and got 16.9 for the for loops code and 3.54 for the vectorized code.

I don't know why there would be such a discrepancy.  This may sound dumb,
but are you sure you didn't confuse the two times?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090914/6ebe439a/attachment.html 

From highegg at gmail.com  Mon Sep 14 12:46:32 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 14 Sep 2009 19:46:32 +0200
Subject: octave vs. matlab vectorization
In-Reply-To: <6800c3890909141043g32e6891ds4bde74e8bc78ec71@mail.gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<19118.32144.109057.627543@segfault.lan>
	<6800c3890909141043g32e6891ds4bde74e8bc78ec71@mail.gmail.com>
Message-ID: <69d8d540909141046r75cef779g7829dad3ab41b571@mail.gmail.com>

On Mon, Sep 14, 2009 at 7:43 PM, James Sherman Jr. <shermanj at umd.edu> wrote:
>>ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
>>
>>i find this result quite staggering.
>
> I ran those two codes with the matlab I have on my laptop (7.7.0.471 R2008b)
> and got 16.9 for the for loops code and 3.54 for the vectorized code.
>
> I don't know why there would be such a discrepancy.? This may sound dumb,
> but are you sure you didn't confuse the two times?
>

IIRC, it may make a difference whether the code is in a
function/script/entered from cmd-line. Functions are best optimized, I
think.


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From highegg at gmail.com  Mon Sep 14 12:51:43 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 14 Sep 2009 19:51:43 +0200
Subject: octave vs. matlab vectorization
In-Reply-To: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
Message-ID: <69d8d540909141051lbbf3030qe7d53fe8bdd1f4f4@mail.gmail.com>

On Mon, Sep 14, 2009 at 5:33 PM, matt <mangostein17 at googlemail.com> wrote:
> i have used octave exclusively for a number of years and just
> recently had to use matlab for teaching purposes.
>
> since i was raised to vectorize loop structures whenever possible, i
> thought i would illustrate the the speed-up for the students in the
> class by comparing execution time for two codes--code 1 with loops,
> and code 2 vectorized.
>
> here is code 1:
>
> t0=clock;
> for m=1:1000000
> ? ? ? ?q=zeros(1,10);
> ? ? ? ?for k=1:10,
> ? ? ? ? ? ? ? ?tmp=k^2;
> ? ? ? ? ? ? ? ?if tmp>50,
> ? ? ? ? ? ? ? ? ? ? ? ?q(k)=tmp;
> ? ? ? ? ? ? ? ?else
> ? ? ? ? ? ? ? ? ? ? ? ?q(k)=0;
> ? ? ? ? ? ? ? ?end
> ? ? ? ?end
> end
> etime(clock,t0)
>
> here is code 2:
>
> t0=clock;
> for m=1:1000000
> ? ? ? ?q=zeros(1,10);
> ? ? ? ?tmp=(1:10).^2;
> ? ? ? ?q=tmp.*(tmp>50);
> end
> etime(clock,t0)
>
> on my laptop, octave (3.0.5) takes 200 secs for the first code and
> just 24 secs for the second--i thought i had proved my point.
>
> however, on the department server (admittedly a much faster machine),
> matlab ground through the first code in 0.82 secs while the second
> code took 4.3 secs.
>
> so, my question: has matlab been optimized for loops? is there some
> sort of compilation going on? any insight gratefully received. i have
> been writing code in octave for a while now and had assumed it would
> almost be a straight port to matlab.
>

Octave mailing list is not the best place to ask technical questions
about Matlab.
But yes, one of the primary performance differences between Octave and
Matlab is that Octave is purely interpreted, while Matlab is JIT
compiled. I think this situation is unlikely to change in the near
future.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From Potorti at isti.cnr.it  Mon Sep 14 15:12:35 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Mon, 14 Sep 2009 22:12:35 +0200
Subject: octave vs. matlab vectorization
In-Reply-To: <BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
Message-ID: <E1MnHud-0004Io-49@tucano.isti.cnr.it>

>ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
>i find this result quite staggering.

Mh.  That vectorised code is slower than jit compiled code is strange,
indeed.  If your measurements are correct, this may depend on various
things, but these are Matlab's implementation issues.

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From abce at Live.cn  Mon Sep 14 21:25:25 2009
From: abce at Live.cn (flyrain)
Date: Mon, 14 Sep 2009 19:25:25 -0700 (PDT)
Subject: unofficial octave 3.2.2 msvc2008 solution/project files
In-Reply-To: <25368142.post@talk.nabble.com>
References: <25368142.post@talk.nabble.com>
Message-ID: <25446848.post@talk.nabble.com>


Hi choonway
*****************************************
1. download and unzip sources octave-3.2.2.tar.bz2
2. move the folder build_vc2008 inside the source folder
*****************************************
does the point 2 means I should create a new folder named source and then
move the build_vs2008 inside? 
there are some build errors..
******************************
2>Generating Code...
2>Build log was saved at
"file://f:\Library\octave-3.2.2-msvc9-20090909.tar\octave-3.2.2-msvc9-20090909\octave-3.2.2\build_vs2008\libcruft\Debug\BuildLog.htm"
2>libcruft - 3 error(s), 2865 warning(s)
3>------ Build started: Project: liboctave, Configuration: Debug Win32
------
3>Compiling...
3>Array-b.cc
3>c1xx : fatal error C1083: Cannot open source file:
'..\..\liboctave\Array-b.cc': No such file or directory
3>Array-C.cc
3>c1xx : fatal error C1083: Cannot open source file:
'..\..\liboctave\Array-C.cc': No such file or directory
3>Array-ch.cc
3>c1xx : fatal error C1083: Cannot open source file:
'..\..\liboctave\Array-ch.cc': No such file or directory
3>Array-d.cc
3>c1xx : fatal error C1083: Cannot open source file:
'..\..\liboctave\Array-d.cc': No such file or directory
3>Array-f.cc
3>c1xx : fatal error C1083: Cannot open source file:
'..\..\liboctave\Array-f.cc': No such file or directory
3>Array-fC.cc
******************************

Liao Choon Way wrote:
> 
> hi all,
> 
> i have prepared a _rough_ set of files for those interested in embedding
> octave
> into windows applications and have problems with mingw dlls.
> 
> Get them 
> http://sourceforge.net/projects/el-rekka/files/octave-3.2.2-msvc9-20090909.tar.gz/download
> here .
> 
> instructions enclosed. read before use. check the repository for updates.
> 
> choonway
> 

-- 
View this message in context: http://www.nabble.com/unofficial-octave-3.2.2-msvc2008-solution-project-files-tp25368142p25446848.html
Sent from the Octave - General mailing list archive at Nabble.com.


From highegg at gmail.com  Tue Sep 15 01:37:36 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Tue, 15 Sep 2009 08:37:36 +0200
Subject: octave vs. matlab vectorization
In-Reply-To: <E1MnHud-0004Io-49@tucano.isti.cnr.it>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<E1MnHud-0004Io-49@tucano.isti.cnr.it>
Message-ID: <69d8d540909142337o51b620b8sa76d3f48aa178200@mail.gmail.com>

On Mon, Sep 14, 2009 at 10:12 PM, Francesco Potorti`
<Potorti at isti.cnr.it> wrote:
>>ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
>>i find this result quite staggering.
>
> Mh. ?That vectorised code is slower than jit compiled code is strange,
> indeed. ?If your measurements are correct, this may depend on various
> things, but these are Matlab's implementation issues.
>

Not necessarily. Compiled loops will often be somewhat faster than
equivalent vectorized code. Look at the original example - in the
loopy code, there's two inner loops; one is zeros() and one the fill.
Compared to that, the semi-vectorized case comprises three loops,
which the JIT compiler may or may not (apparently not in this case) be
able to merge.

I intentionally wrote "semi-vectorized" because, from Octave's point
of view, neither code is vectorized. I'm sure that in Octave, the
"interpreter overhead" plays the major part in *both* cases. Such
example is just unrealistic for real-life coding in Octave, because it
actually failed to vectorize anything.
Vectorizing an inner loop often helps, but not when the ratio
inner:outer number of cycles is 1:1e6. To get good performance,
generally the ratio inner:outer should be at most 1 in terms of orders
of magnitude.

A final thought about the JIT compilation:
Imprecisely quoting MathWorks site, "MathWorks is committed to make
Matlab as fast as traditional HPC languages such as C, C++ and
Fortran". Besides the general unreasonability of the statement
(comparing speed of languages is usually a nonsense),
I consider this a clever marketing strategy. MathWorks surely realizes
that the MEX interface sucks. I think there were number of third-party
attempts to create an OOP interface to the Matlab engine, but
MathWorks adopted none, neither they invented their own one. And for
good reason, I think - improving interoperability with C/C++/Fortran
could make customers to invest in compilers and better C/C++/Fortran
programming skills. On the contrary, the situation where everything is
written in the Matlab language is much more favorable for the company
- in case your code is not fast enough, your best option is to buy a
newer Matlab and hope the JIT compiler has improved enough to do what
you meant to do. I think Matlab comes with little documentation on
what exactly the JIT compiler can handle, let alone any guarantees, so
you're entirely left at its mercy. In short, Matlab is now improving
in the area where C and Fortran compilers have been shining for
decades. The Matlab language was designed to be interpreted, and is
almost uniquely ill-suited for compilation. Despite that, the company
made remarkable progress in teaching the elephant to dance. Of course,
MathWorks needed some way forward, or they would lose revenues, and
they picked up this one. But that does not necessarily mean Octave
will follow. I daresay that most of those who actively contribute to
Octave *can* speak C, C++ or Fortran. And people will not stop using
C, C++ and Fortran anywhere in near future. So I don't reckon Octave
will JIT-compile in near future; I think there's just not enough
interest. Still, I might be wrong.

regards

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From choonway at gmail.com  Tue Sep 15 02:04:28 2009
From: choonway at gmail.com (Liao Choon Way)
Date: Tue, 15 Sep 2009 00:04:28 -0700 (PDT)
Subject: unofficial octave 3.2.2 msvc2008 solution/project files
In-Reply-To: <25446848.post@talk.nabble.com>
References: <25368142.post@talk.nabble.com> <25446848.post@talk.nabble.com>
Message-ID: <25449018.post@talk.nabble.com>


Hi,

Oops. the folder should be in the root directory, where the README.* files
are.

Apologies,

choonway.



flyrain wrote:
> 
> Hi choonway
> *****************************************
> 1. download and unzip sources octave-3.2.2.tar.bz2
> 2. move the folder build_vc2008 inside the source folder
> *****************************************
> does the point 2 means I should create a new folder named source and then
> move the build_vs2008 inside? 
> there are some build errors..
> ******************************
> 2>Generating Code...
> 2>Build log was saved at
> "file://f:\Library\octave-3.2.2-msvc9-20090909.tar\octave-3.2.2-msvc9-20090909\octave-3.2.2\build_vs2008\libcruft\Debug\BuildLog.htm"
> 2>libcruft - 3 error(s), 2865 warning(s)
> 3>------ Build started: Project: liboctave, Configuration: Debug Win32
> ------
> 3>Compiling...
> 3>Array-b.cc
> 3>c1xx : fatal error C1083: Cannot open source file:
> '..\..\liboctave\Array-b.cc': No such file or directory
> 3>Array-C.cc
> 3>c1xx : fatal error C1083: Cannot open source file:
> '..\..\liboctave\Array-C.cc': No such file or directory
> 3>Array-ch.cc
> 3>c1xx : fatal error C1083: Cannot open source file:
> '..\..\liboctave\Array-ch.cc': No such file or directory
> 3>Array-d.cc
> 3>c1xx : fatal error C1083: Cannot open source file:
> '..\..\liboctave\Array-d.cc': No such file or directory
> 3>Array-f.cc
> 3>c1xx : fatal error C1083: Cannot open source file:
> '..\..\liboctave\Array-f.cc': No such file or directory
> 3>Array-fC.cc
> ******************************
> 
> Liao Choon Way wrote:
>> 
>> hi all,
>> 
>> i have prepared a _rough_ set of files for those interested in embedding
>> octave
>> into windows applications and have problems with mingw dlls.
>> 
>> Get them 
>> http://sourceforge.net/projects/el-rekka/files/octave-3.2.2-msvc9-20090909.tar.gz/download
>> here .
>> 
>> instructions enclosed. read before use. check the repository for updates.
>> 
>> choonway
>> 
> 
> 

-- 
View this message in context: http://www.nabble.com/unofficial-octave-3.2.2-msvc2008-solution-project-files-tp25368142p25449018.html
Sent from the Octave - General mailing list archive at Nabble.com.


From lgjsheron at gmail.com  Tue Sep 15 01:41:46 2009
From: lgjsheron at gmail.com (Sheron Gerard Jeshuran)
Date: Tue, 15 Sep 2009 12:41:46 +0600
Subject: Linespace
Message-ID: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>

Dear SIr,

I have a problem to draw a graph in Octave my function is f(x)=4-3x and also
i need f inverse also, and new and the time i run the command following
error message comes

octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> linespace(0, 2*pi, 100);
error: `linespace' undefined near line 1 column 1
octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
>


-- 
Sheron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090915/5612f6d7/attachment-0001.html 

From mangostein17 at googlemail.com  Tue Sep 15 03:58:21 2009
From: mangostein17 at googlemail.com (matt)
Date: Tue, 15 Sep 2009 09:58:21 +0100
Subject: octave vs. matlab vectorization
In-Reply-To: <6800c3890909141043g32e6891ds4bde74e8bc78ec71@mail.gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<19118.32144.109057.627543@segfault.lan>
	<6800c3890909141043g32e6891ds4bde74e8bc78ec71@mail.gmail.com>
Message-ID: <17E8A775-B0BD-4A2F-BB8B-DD1D09C6B6C0@gmail.com>


hi james,

good point! i have rechecked and the loop code *is* 5x quicker--this  
with matlab 7.5.0.338 and, i just noticed, matlab compiler 4.7.


On 14 Sep 2009, at 18:43, James Sherman Jr. wrote:

> >ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
> >
> >i find this result quite staggering.
>
> I ran those two codes with the matlab I have on my laptop  
> (7.7.0.471 R2008b) and got 16.9 for the for loops code and 3.54 for  
> the vectorized code.
>
> I don't know why there would be such a discrepancy.  This may sound  
> dumb, but are you sure you didn't confuse the two times?


From mangostein17 at googlemail.com  Tue Sep 15 04:28:35 2009
From: mangostein17 at googlemail.com (matt)
Date: Tue, 15 Sep 2009 10:28:35 +0100
Subject: octave vs. matlab vectorization
In-Reply-To: <19118.32144.109057.627543@segfault.lan>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<19118.32144.109057.627543@segfault.lan>
Message-ID: <704BE6F4-E39F-40C5-B201-EDF8D97D5E6A@gmail.com>


Hi John,

First let me take this opportunity to thank you for all the great  
work you have done on Octave over the years. I am using it constantly  
and couldn't be happier.

On vectorization issue, I was staggered for two reasons. First, I  
couldn't believe that Matlab had "given up" on software  
vectorization. Did they tell anyone?! I only really care because I  
sometimes send Octave code to Matlab users; it's nice to know what  
the differences are. The second reason is that I was surprised the  
JIT compiler worked as well as it did--to use Jaroslav's quote, the  
elephant can dance.

Having said that, James found the vectorized code ran 5x faster on  
his Matlab--so who knows what is going on? Perhaps the Matlab  
compiler is cheating?

Anyway, many thanks again for Octave.

Best,

Matt.

On 14 Sep 2009, at 18:29, John W. Eaton wrote:

> On 14-Sep-2009, matt wrote:
>
> | ah, no! code 1 was *faster* under matlab! the ratio is 0.19:1
> |
> | i find this result quite staggering.
>
> Why?  As someone else already explained, Matlab has a JIT compiler
> that can, in some cases at least, compile loops into native code so
> that it runs faster and Octave does not have a JIT compiler.
>
> jwe


From mangostein17 at googlemail.com  Tue Sep 15 07:01:05 2009
From: mangostein17 at googlemail.com (matt)
Date: Tue, 15 Sep 2009 13:01:05 +0100
Subject: octave vs. matlab vectorization
In-Reply-To: <69d8d540909142337o51b620b8sa76d3f48aa178200@mail.gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<E1MnHud-0004Io-49@tucano.isti.cnr.it>
	<69d8d540909142337o51b620b8sa76d3f48aa178200@mail.gmail.com>
Message-ID: <6CAC17DF-3E81-4E6F-9E5C-2CE07B3F847D@gmail.com>


thanks for the insight on compiler issues.

the matlab JIT compiler is doing something clever in the simple- 
minded codes i was experimenting with but i realise now that a lot of  
the problem is the overhead in allocating and defining arrays.

i had a vector q of length 10. in the loops code, it was fast to  
check through it elementwise; in the (semi) vectorized code, time was  
being spent setting up q and also tmp and (tmp>50), both length 10.

replacing 10 by 1000 puts the two codes on a par as now the overhead  
to set up the vectors is relatively small.

in octave, of course, the vectorized version wins by miles. 

From soren at hauberg.org  Tue Sep 15 08:22:14 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Tue, 15 Sep 2009 15:22:14 +0200
Subject: Linespace
In-Reply-To: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>
References: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>
Message-ID: <1253020934.14685.0.camel@sh-t400>

tir, 15 09 2009 kl. 12:41 +0600, skrev Sheron Gerard Jeshuran:
> Dear SIr,
> 
> I have a problem to draw a graph in Octave my function is f(x)=4-3x
> and also i need f inverse also, and new and the time i run the command
> following error message comes
> 
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> > linespace(0, 2*pi, 100);
> error: `linespace' undefined near line 1 column 1
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin

The function you want is called 'linspace'


From bpabbott at mac.com  Tue Sep 15 08:23:01 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Tue, 15 Sep 2009 09:23:01 -0400
Subject: Linespace
In-Reply-To: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>
References: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>
Message-ID: <7A60E532-D8B1-4FD2-AEF5-989919967A3E@mac.com>

On Sep 15, 2009, at 2:41 AM, Sheron Gerard Jeshuran  
<lgjsheron at gmail.com> wrote:

> Dear SIr,
>
> I have a problem to draw a graph in Octave my function is f(x)=4-3x  
> and also i need f inverse also, and new and the time i run the  
> command following error message comes
>
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> > linespace(0, 2*pi, 100);
> error: `linespace' undefined near line 1 column 1
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> >

Try "linspace" instead of "linespace".

Ben

From storrsjm at email.uc.edu  Tue Sep 15 10:32:23 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Tue, 15 Sep 2009 11:32:23 -0400
Subject: octave vs. matlab vectorization
In-Reply-To: <704BE6F4-E39F-40C5-B201-EDF8D97D5E6A@gmail.com>
References: <343F894B-E4BD-43F7-9F2A-9383829E302C@gmail.com>
	<E1MnESl-0002Bd-1r@tucano.isti.cnr.it>
	<BC507FCE-B0DE-49F1-8F11-26FBC5FAB300@gmail.com>
	<19118.32144.109057.627543@segfault.lan>
	<704BE6F4-E39F-40C5-B201-EDF8D97D5E6A@gmail.com>
Message-ID: <a0daad600909150832k70aa2cd5y9db437b6c64b58c1@mail.gmail.com>

On Tue, Sep 15, 2009 at 5:28 AM, matt <mangostein17 at googlemail.com> wrote:

> Having said that, James found the vectorized code ran 5x faster on
> his Matlab--so who knows what is going on? Perhaps the Matlab
> compiler is cheating?
>

I wouldn't call working well cheating ;)

As it was explained to me when 7.0 came out, the JIT compiler is able to
detect certain idioms and to replace them with very optimized code. In the
7.0 days it was pretty hit or miss whether the JIT optimizations would kick
in or not and I remember at a training session that the application engineer
from Mathworks described in great detail how to monkey around with loop
structure and variable use in order to help the JIT detect the idioms. In
the following years I imagine their idiom detection has improved
substantially.

--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090915/c1c9fb17/attachment.html 

From arnoques at gmail.com  Mon Sep 14 20:16:29 2009
From: arnoques at gmail.com (Pablo)
Date: Mon, 14 Sep 2009 21:16:29 -0400
Subject: Randomly generate 10 different integers
In-Reply-To: <E971ED63-4931-40BA-9EA8-199539460F10@utk.edu>
References: <7a394d4e0906301045v2c1f13b9gdff231dcac4cdd6@mail.gmail.com>
	<E971ED63-4931-40BA-9EA8-199539460F10@utk.edu>
Message-ID: <4AAEEAED.3020308@gmail.com>

Rob Mahurin wrote:
> On Jun 30, 2009, at 1:45 PM, Xin Dong wrote:
>> I wanna randomly generate 10 different integers in a given range. I  
>> know I can use random function and discard duplicate values. But is  
>> there any other simple way to do it?
> 
> One way is
> 
> 	[junk, index] = sort(rand(1,10));
> 	disp(index)
> 
> Rob
> 

I had to do this recently. The above example just gives a random
permutation of the numbers 1 to 10 (the randperm function does exactly
that). To get N random numbers from dmin to dmax you sholud do:
D = randperm(dmax - dmin + 1);
D = D(:,1:N);
D = D + dmin - 1;

This is not very efficient if you need to select very few numbers, or
almost all, from a very large range, because I was not interested in the
order of the numbers, and you have to sort every number of the range.

If you have a very large range (I used randperm(10000), and it gets VERY
slow!) you might replace the above with a custom sort function, that
only sorts the first N numbers. That way, you can shave off a LOT of time.

I know this comes a little late, but it might still help somebody!

  Arnoques

From donwalkertt at hotmail.com  Tue Sep 15 12:27:10 2009
From: donwalkertt at hotmail.com (donwalker)
Date: Tue, 15 Sep 2009 10:27:10 -0700 (PDT)
Subject: LSODE-opts.h
Message-ID: <25458273.post@talk.nabble.com>


// Generated automatically from LSODE-opts.in.
...
#include <ODE.h>
...

I am using a tutorial I found for using octave functions in C++. One of the
examples uses LSODE-opts.h. The line above generates an error. Shouldn't it
be 

#include "ODE.h"

?

Thanks for help from anyone


-- 
View this message in context: http://www.nabble.com/LSODE-opts.h-tp25458273p25458273.html
Sent from the Octave - General mailing list archive at Nabble.com.


From tmacchant at yahoo.co.jp  Tue Sep 15 13:33:45 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Wed, 16 Sep 2009 03:33:45 +0900 (JST)
Subject: Linespace
In-Reply-To: <72ee9c210909142341h31615ff1nf87a9bb84c749706@mail.gmail.com>
Message-ID: <20090915183345.63295.qmail@web3814.mail.bbt.yahoo.co.jp>

Hello

I guess that 'lin' means 'linear' so that 'lin' is proper than 'line' as  an abbreviation.
For logarithmic scale, the 'logspace' is used.

Regards

Tatsuro
--- Sheron Gerard Jeshuranwrote:

> Dear SIr,
> 
> I have a problem to draw a graph in Octave my function is f(x)=4-3x and also
> i need f inverse also, and new and the time i run the command following
> error message comes
> 
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> > linespace(0, 2*pi, 100);
> error: `linespace' undefined near line 1 column 1
> octave-3.2.2.exe:1:C:\Octave\3.2.2_gcc-4.3.0\bin
> >
> 
> 
> -- 
> Sheron
> > _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From fl at ifi.uio.no  Tue Sep 15 14:20:22 2009
From: fl at ifi.uio.no (Fredrik Lingvall)
Date: Tue, 15 Sep 2009 21:20:22 +0200
Subject: avifile (and all that jazz)
In-Reply-To: <887085.21640.qm@web30205.mail.mud.yahoo.com>
References: <887085.21640.qm@web30205.mail.mud.yahoo.com>
Message-ID: <355366570909151220i340b68b5sffb8ebcd9ea1929c@mail.gmail.com>

You can do this by generating a series of, for example, png-images and then
let mencoder put them together to an avi-file. Something like:

png_path = 'png/';

for k=1:K

     % Plot the k:th image here.
     image(A(:,:,k));

     if k<10
        eval(['print("' png_path 'snapshot_000' num2str(k) '.png",
"-dpng");']);
      elseif k<100
        eval(['print("' png_path 'snapshot_00' num2str(k) '.png",
"-dpng");']);
      elseif k<1000
        eval(['print("' png_path 'snapshot_0' num2str(k) '.png",
"-dpng");']);
      else
        eval(['print("' png_path 'snapshot_' num2str(k) '.png",
"-dpng");']);
      end
end

system(['mencoder -quiet mf://png/*.png -mf w=640:h=480:fps=' ...
            '25:type=png -ovc lavc -lavcopts vcodec=wmv1 -oac copy -' ...
            'o snapshots.avi']);

/Fredrik


On Fri, Sep 11, 2009 at 8:19 PM, Kristen Richter <christy1865 at yahoo.com>wrote:

> Dear helpful mailing list,
>
> I am attempting to turn a series of plots into an avifile.  While I've
> installed the video add-on package from Octave-forge, I can't add frames to
> my .avi because the AVHandler can't seem to find the appropriate codecs.
>  Using the example below, found here:
> http://neacm.fe.up.pt/pub/octave/MAILING-LISTS/help-octave/2005/3694
>
> # Open the AVI file
> m = avifile("test.avi", "codec", "msmpeg4v2")
>
> # Generate and add frames
> for i = 1:100
>  I = zeros(100,100);
>  I(i,:) = i;
>  I(:,i) = 200-i;
>  addframe(m, I/255)
>  printf(".")
> endfor
> printf("\n")
>
> # Close the file
> clear mI get the following error:
>
> AVHandler: cannot find codec msmpeg4v2
> error: avifile: AVHandler setup failed
> addframe: error adding frame to avifile
> I've tried the example, specifying many different codecs, including
> cinepak, indeo5, and indeo3.  I've checked my synaptic package manager for
> appropriate codecs... currently I've installed libavifile (vs. 0.7), which
> mentions the indeo library by name.
>
> Are the codecs supposed to be installed in a specific directory? Has anyone
> come across similar issues in the past?
> The example is pretty old... dated 2005, and I know the video package was
> updated last this past May.
>
> Any insight on the situation would be greatly appreciated. As always, your
> help in the past has been invaluable.
>
> -Kristen Richter
>
> By the way, does anyone know of a good work around for the getframe
> function?... I have yet to think of a good way to capture the figure on the
> screen into an appropriately formatted image matrix.
>
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090915/c7dab22c/attachment.html 

From soren at hauberg.org  Tue Sep 15 14:26:20 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Tue, 15 Sep 2009 21:26:20 +0200
Subject: avifile (and all that jazz)
In-Reply-To: <355366570909151220i340b68b5sffb8ebcd9ea1929c@mail.gmail.com>
References: <887085.21640.qm@web30205.mail.mud.yahoo.com>
	<355366570909151220i340b68b5sffb8ebcd9ea1929c@mail.gmail.com>
Message-ID: <1253042780.4603.5.camel@sh-t400>

I believe this block

>      if k<10
>         eval(['print("' png_path 'snapshot_000' num2str(k) '.png",
> "-dpng");']); 
>       elseif k<100
>         eval(['print("' png_path 'snapshot_00' num2str(k) '.png",
> "-dpng");']); 
>       elseif k<1000
>         eval(['print("' png_path 'snapshot_0' num2str(k) '.png",
> "-dpng");']); 
>       else
>         eval(['print("' png_path 'snapshot_' num2str(k) '.png",
> "-dpng");']); 
>       end

can be replaced by

  print (sprintf ("%s/snapshot_%.5d.png", png_path, k));

but I haven't tested it, so it might need slight alterations. I
principle you should be using 'fullfile' concatenate the directory name
with the file name as that would give you platform independence, but I
didn't bother doing that here.

S?ren


From estebanvoehringer at hotmail.com  Tue Sep 15 15:14:25 2009
From: estebanvoehringer at hotmail.com (esteban1)
Date: Tue, 15 Sep 2009 13:14:25 -0700 (PDT)
Subject: regdatasmooth
Message-ID: <25461009.post@talk.nabble.com>


Hi,

I have a problem with the regdatasmooth function implemented in the
data-smoothing-1.2.0 package. Whenever I want to run a demo or the function
itself I get the following error: 
error: unary operator `'' not implemented for `list' operands.

The regdatasmooth function calls the rgdtsmcore function. This function
works fine, when I run it with my parameters.

But I would like to use regdatasmooth since it optimizes the smoothing
through lambda and other parameters.
I allready searched the web and the mailing list but I was not able to find
something.

Thanks for the help!
-- 
View this message in context: http://www.nabble.com/regdatasmooth-tp25461009p25461009.html
Sent from the Octave - General mailing list archive at Nabble.com.


From sttt at gmx.de  Tue Sep 15 12:28:07 2009
From: sttt at gmx.de (Steffen Titze)
Date: Tue, 15 Sep 2009 19:28:07 +0200
Subject: Load data with script
Message-ID: <4AAFCEA7.50903@gmx.de>

Hello,

i have written a script to load some values from a data file. With old 
version of octave (3.0.3) i don?t had anyproblems. But with the new 
version i cann?t load the data file with the script.

if i load the data file with same syntax with out the script, i don?t 
have any problems.

x = load ('C:\Test\090811.txt')

Error massage with script
error: invalid call to script

How can i solve the problem? I use a windows pc.

Thanks
Steffen

From jwe at octave.org  Wed Sep 16 02:45:30 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 16 Sep 2009 03:45:30 -0400
Subject: Load data with script
In-Reply-To: <4AAFCEA7.50903@gmx.de>
References: <4AAFCEA7.50903@gmx.de>
Message-ID: <19120.38810.727236.986407@segfault.lan>

On 15-Sep-2009, Steffen Titze wrote:

| i have written a script to load some values from a data file. With old 
| version of octave (3.0.3) i don?t had anyproblems. But with the new 
| version i cann?t load the data file with the script.
| 
| if i load the data file with same syntax with out the script, i don?t 
| have any problems.
| 
| x = load ('C:\Test\090811.txt')
| 
| Error massage with script
| error: invalid call to script
| 
| How can i solve the problem? I use a windows pc.

Is there a load.m file somewhere in Octave's path?  Try

  which load

and see what it tells you.

jwe


From ct529 at york.ac.uk  Wed Sep 16 03:10:37 2009
From: ct529 at york.ac.uk (Corrado)
Date: Wed, 16 Sep 2009 09:10:37 +0100
Subject: What did I do wrong with this funtion? - Implementing msplines and
	isplines in Octave
Message-ID: <200909160910.37821.ct529@york.ac.uk>

Dear friends,

I am trying to implement in Octave the two spline families as in the 
documentation here below attached. I am a newbie to Octave, and this is my 
first function.

I built the function:

function retspline = mspline (i,x,k,t)
	# i is the spline index in the family
	# x is the variable
	# t is the vector of knots, t(h) is the h-th knot
	# k is the Mspline degree
	
	I=i
	
	if(k==1)
		
		if(x<t(i+1) & x>=t(i))
			td=t(i+1)-t(i)				
			M=1/td			
		else
			M=0
		endif

	else	
		
		kk=(k-1)
		
		if (x>=t(i) & x<t(i+k))
			M=k*((x-t(i))*mspline(i=I,x=x,k=kk,t=t)+(t(i+k)-
x)*mspline(i=(I+1),x=x,k=kk,t=t))/((k-1)*(t(i+k)-t(i)))
		elseif (x<t(i) | x>=t(i+k))
			M=0
		endif

	endif
		
	retspline=M	

endfunction

but for some reason it does not work ....

I add that I am a newbie with Octave, please do not shut me.

What did I do wrong?

Best,
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk

-------------- next part --------------
A non-text attachment was scrubbed...
Name: mspline
Type: image/png
Size: 71260 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090916/3421d5ed/attachment-0002.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ispline.png
Type: image/png
Size: 51862 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090916/3421d5ed/attachment-0003.png 

From highegg at gmail.com  Wed Sep 16 04:26:43 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Wed, 16 Sep 2009 11:26:43 +0200
Subject: What did I do wrong with this funtion? - Implementing msplines 
	and isplines in Octave
In-Reply-To: <200909160910.37821.ct529@york.ac.uk>
References: <200909160910.37821.ct529@york.ac.uk>
Message-ID: <69d8d540909160226x9b81ad8l4be77f639b5b1a3b@mail.gmail.com>

On Wed, Sep 16, 2009 at 10:10 AM, Corrado <ct529 at york.ac.uk> wrote:
> Dear friends,
>
> I am trying to implement in Octave the two spline families as in the
> documentation here below attached. I am a newbie to Octave, and this is my
> first function.
>
> I built the function:
>
> function retspline = mspline (i,x,k,t)
> ? ? ? ?# i is the spline index in the family
> ? ? ? ?# x is the variable
> ? ? ? ?# t is the vector of knots, t(h) is the h-th knot
> ? ? ? ?# k is the Mspline degree
>
> ? ? ? ?I=i
>
> ? ? ? ?if(k==1)
>
> ? ? ? ? ? ? ? ?if(x<t(i+1) & x>=t(i))
> ? ? ? ? ? ? ? ? ? ? ? ?td=t(i+1)-t(i)
> ? ? ? ? ? ? ? ? ? ? ? ?M=1/td
> ? ? ? ? ? ? ? ?else
> ? ? ? ? ? ? ? ? ? ? ? ?M=0
> ? ? ? ? ? ? ? ?endif
>
> ? ? ? ?else
>
> ? ? ? ? ? ? ? ?kk=(k-1)
>
> ? ? ? ? ? ? ? ?if (x>=t(i) & x<t(i+k))
> ? ? ? ? ? ? ? ? ? ? ? ?M=k*((x-t(i))*mspline(i=I,x=x,k=kk,t=t)+(t(i+k)-
> x)*mspline(i=(I+1),x=x,k=kk,t=t))/((k-1)*(t(i+k)-t(i)))
> ? ? ? ? ? ? ? ?elseif (x<t(i) | x>=t(i+k))
> ? ? ? ? ? ? ? ? ? ? ? ?M=0
> ? ? ? ? ? ? ? ?endif
>
> ? ? ? ?endif
>
> ? ? ? ?retspline=M
>
> endfunction
>
> but for some reason it does not work ....
>
> I add that I am a newbie with Octave, please do not shut me.
>
> What did I do wrong?
>
> Best,
> --
> Corrado Topi
>

You're probably misinterpreting the expression
`mspline(i=I,x=x,k=kk,t=t)'  (did you program in Python or Fortran?)
In Octave, there is no passing parameters by name, and the expression
is interpreted like in C, so that it means:
i=I,x=x,k=kk,t=t,mspline(i,x,k,t)
which probably spoils your script by overwriting variables. I didn't
check for other problems.
Note that even if you succeed, this will be a very inefficient
implementation in Octave, but probably good enough to start
programming.

hth

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From ct529 at york.ac.uk  Wed Sep 16 04:33:11 2009
From: ct529 at york.ac.uk (Corrado)
Date: Wed, 16 Sep 2009 10:33:11 +0100
Subject: What did I do wrong with this funtion? - Implementing msplines
In-Reply-To: <69d8d540909160226x9b81ad8l4be77f639b5b1a3b@mail.gmail.com>
References: <200909160910.37821.ct529@york.ac.uk>
	<69d8d540909160226x9b81ad8l4be77f639b5b1a3b@mail.gmail.com>
Message-ID: <200909161033.11564.ct529@york.ac.uk>

Dear Jaroslav,

thanks for your help. I prototyped using R originally, and then rewrote it in 
Octave. Two quick questions:

1) How can I make the function more efficient?
2) Do you think I should rather write it in Fortran or any other language, 
instead of Octave?

Regards

On Wednesday 16 September 2009 10:26:43 Jaroslav Hajek wrote:
> On Wed, Sep 16, 2009 at 10:10 AM, Corrado <ct529 at york.ac.uk> wrote:
> > Dear friends,
> >
> > I am trying to implement in Octave the two spline families as in the
> > documentation here below attached. I am a newbie to Octave, and this is
> > my first function.
> >
> > I built the function:
> >
> > function retspline = mspline (i,x,k,t)
> >        # i is the spline index in the family
> >        # x is the variable
> >        # t is the vector of knots, t(h) is the h-th knot
> >        # k is the Mspline degree
> >
> >        I=i
> >
> >        if(k==1)
> >
> >                if(x<t(i+1) & x>=t(i))
> >                        td=t(i+1)-t(i)
> >                        M=1/td
> >                else
> >                        M=0
> >                endif
> >
> >        else
> >
> >                kk=(k-1)
> >
> >                if (x>=t(i) & x<t(i+k))
> >                        M=k*((x-t(i))*mspline(i=I,x=x,k=kk,t=t)+(t(i+k)-
> > x)*mspline(i=(I+1),x=x,k=kk,t=t))/((k-1)*(t(i+k)-t(i)))
> >                elseif (x<t(i) | x>=t(i+k))
> >                        M=0
> >                endif
> >
> >        endif
> >
> >        retspline=M
> >
> > endfunction
> >
> > but for some reason it does not work ....
> >
> > I add that I am a newbie with Octave, please do not shut me.
> >
> > What did I do wrong?
> >
> > Best,
> > --
> > Corrado Topi
>
> You're probably misinterpreting the expression
> `mspline(i=I,x=x,k=kk,t=t)'  (did you program in Python or Fortran?)
> In Octave, there is no passing parameters by name, and the expression
> is interpreted like in C, so that it means:
> i=I,x=x,k=kk,t=t,mspline(i,x,k,t)
> which probably spoils your script by overwriting variables. I didn't
> check for other problems.
> Note that even if you succeed, this will be a very inefficient
> implementation in Octave, but probably good enough to start
> programming.
>
> hth



-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Wed Sep 16 04:45:27 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Wed, 16 Sep 2009 11:45:27 +0200
Subject: What did I do wrong with this funtion? - Implementing msplines
In-Reply-To: <200909161033.11564.ct529@york.ac.uk>
References: <200909160910.37821.ct529@york.ac.uk>
	<69d8d540909160226x9b81ad8l4be77f639b5b1a3b@mail.gmail.com>
	<200909161033.11564.ct529@york.ac.uk>
Message-ID: <69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>

On Wed, Sep 16, 2009 at 11:33 AM, Corrado <ct529 at york.ac.uk> wrote:
> Dear Jaroslav,
>
> thanks for your help. I prototyped using R originally, and then rewrote it in
> Octave. Two quick questions:
>
> 1) How can I make the function more efficient?

I can hardly give advices given my little knowledge about your
problem, but it seems to me that M forms a sort of divided-differences
table, which is common with splines. In such case, it is much better
to build the table using loops (this is true for any language) and
vectorize the inner loops (this is for Octave).

> 2) Do you think I should rather write it in Fortran or any other language,
> instead of Octave?
>

With Fortran or C++ you can often (but not always) achieve a higher
performance, but my common practice is write in Octave first, debug
etc, and possibly recode later if more performance is needed. Octave
is just ideal for that sort of work.

Btw. please DON'T change the subject of the topic arbitrarily, it
confuses mailers.

hth

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Wed Sep 16 04:53:21 2009
From: ct529 at york.ac.uk (Corrado)
Date: Wed, 16 Sep 2009 10:53:21 +0100
Subject: What did I do wrong with this funtion? - Implementing msplines
In-Reply-To: <69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>
References: <200909160910.37821.ct529@york.ac.uk>
	<200909161033.11564.ct529@york.ac.uk>
	<69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>
Message-ID: <200909161053.21313.ct529@york.ac.uk>

On Wednesday 16 September 2009 10:45:27 Jaroslav Hajek wrote:
> Btw. please DON'T change the subject of the topic arbitrarily, it
> confuses mailers.

I did not change the subject at all .... why are you saying that?

-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Wed Sep 16 04:56:53 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Wed, 16 Sep 2009 11:56:53 +0200
Subject: What did I do wrong with this funtion? - Implementing msplines
In-Reply-To: <200909161053.21313.ct529@york.ac.uk>
References: <200909160910.37821.ct529@york.ac.uk>
	<200909161033.11564.ct529@york.ac.uk>
	<69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>
	<200909161053.21313.ct529@york.ac.uk>
Message-ID: <69d8d540909160256n21a2bc96w239b9cb54a1de42e@mail.gmail.com>

On Wed, Sep 16, 2009 at 11:53 AM, Corrado <ct529 at york.ac.uk> wrote:
> On Wednesday 16 September 2009 10:45:27 Jaroslav Hajek wrote:
>> Btw. please DON'T change the subject of the topic arbitrarily, it
>> confuses mailers.
>
> I did not change the subject at all .... why are you saying that?
>
> --
> Corrado Topi
>
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>
>

Your first message was
"What did I do wrong with this funtion? - Implementing msplines and
isplines in Octave"
but your subsequent replies had the subject
"Re: What did I do wrong with this funtion? - Implementing msplines"
(trailing words truncated). If this happened without your intention,
maybe your mailer is misconfigured, but it surely isn't a good idea.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Wed Sep 16 05:38:36 2009
From: ct529 at york.ac.uk (Corrado)
Date: Wed, 16 Sep 2009 11:38:36 +0100
Subject: What did I do wrong with this funtion? - Implementing msplines
In-Reply-To: <69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>
References: <200909160910.37821.ct529@york.ac.uk>
	<200909161033.11564.ct529@york.ac.uk>
	<69d8d540909160245i5ee2d7d8y17b89cd01587cdd9@mail.gmail.com>
Message-ID: <200909161138.36444.ct529@york.ac.uk>

Dear Octave list,

I reached a point where the Octave prototype of the mspline function works 
really well, and produces the desired output. The final function can be 
downloaded from here:

http://scsys.co.uk:8002/33893

And the bibliographic reference is attached to the email in jpeg image format 
(mspline_.jpeg).

Now the mspline(i,x,k,t) function returns the correct value when I pass a real 
number on as x.

For example:

t=[0,0,0,.3,.5,.6,1,1,1]
mspline(1,.02,3,t)

returns the correct value.

I would now like to optimise it so that if I pass as x a vector of values or a 
matrix, the function does transform the vector / matrix (that is all values in 
them) in an efficient way. The idea is to create a "for" cycle that spans the 
entire vector or matrix, which calls the mspline function and applies it to 
each element of the vector or matrix. 

Is that a good idea? Any alternative suggestion? 

I particular: 1) What language to use for higher performance (Octave, Fortran, 
....) 2) How to change the structure of the function so that it is better 
performing in Octave.

Please remember I am an Octave newbie, and this is my first attempt at writing 
a package. The aim of the package is to provide mspline and ispline based 
regressions on large size distance matrices to Octave.

Regards
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk

-------------- next part --------------
A non-text attachment was scrubbed...
Name: mspline_.jpeg
Type: image/jpeg
Size: 18384 bytes
Desc: not available
Url : https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090916/df8b47b6/attachment-0001.jpeg 

From Potorti at isti.cnr.it  Wed Sep 16 07:43:45 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Wed, 16 Sep 2009 14:43:45 +0200
Subject: avifile (and all that jazz)
In-Reply-To: <355366570909151220i340b68b5sffb8ebcd9ea1929c@mail.gmail.com>
References: <887085.21640.qm@web30205.mail.mud.yahoo.com>
	<355366570909151220i340b68b5sffb8ebcd9ea1929c@mail.gmail.com>
Message-ID: <E1MntrN-0006Cj-Ui@tucano.isti.cnr.it>

>You can do this by generating a series of, for example, png-images and then
>let mencoder put them together to an avi-file.

Or else use this function of mine.  By the way, wahat people think:
would it be reasonable to add it to the video package?

function movie (action, mmov="octave_movie.mp4")
  ## Create a movie from plots
  ##
  ## Example usage:
  ##   figure("visible","off"); movie("init")
  ##   a=zeros(100,100); a(1:20,41:60)=1;
  ##   for i=1:100; a=shift(a,1); imshow(a); movie("add"); endfor
  ##   movie("close"); close; system("totem octave_movie.mp4")
  ##
  ## There are three commands used as first argment: "init", "add",
  ## "close". By default, a movie named "octave_movie.mp4" is created
  ## using ffmpeg.  An optional second argument allows one changing both
  ## the name and the type of movie.  The type ".dir" creates a
  ## directory containing a png file per frame.  The type ".zip"
  ## archives it using zip.  Types ".mp4", ".ogg", ".mov", ".mjpeg",
  ## ".avi", ".flv" are created using ffmpeg; types ".mng", ".gif" are
  ## created using convert; type ".swf" is created using png2swf.  You
  ## must have the relevant program installed to create a movie with the
  ## corresponding extension; no program is required for ".dir".

  ## Francesco Potort?, 2008
  ## Revision 1.10
  ## License: GPL version 3 or later

  verbose = false;
  rate = 5;			# frames per second

  actions = {"init" "add" "close"};
  # gif swf
  types = {".mp4" ".mng" ".gif" ".zip" ".ogg" ".swf" ".mov" ...
	   ".mjpeg" ".avi" ".flv" ".dir"};
  if (nargin < 1 || !ischar(action) || !any(strcmp(action, actions)))
    error("first argument must be one of:%s", sprintf(" %s",actions{:}));
  endif
  if (nargin != 2 || !ischar(mmov))
    error("second arg must be a string");
  endif
  [mpath mname mtype] = fileparts(mmov);

  mdir = fullfile(mpath, [mname ".d"]);
  ppat =  "%06d.png";
  mpat = fullfile(mdir, ppat);
  mglob = fullfile(mdir, strrep(sprintf(ppat,0),"0","[0-9]"));
  fnof = fullfile(mdir, "+frame-number+");

  switch (action)
    case actions{1}		# init a movie
      if (isdir(mmov))
	cleandir(mmov, verbose)
      else
	unlink(mmov);
      endif
      while (!([allgood msg] = mkdir(mdir)))
	if (stat(fnof) && load(fnof).frameno == 0)
	  error("while creating dir '%s': %s", mdir, msg);
	else
	  cleandir(mdir, verbose);
	endif
      endwhile
      frameno = 0; save("-text",fnof,"frameno");
      if (verbose) printf("Directory '%s' created.\n", mdir); endif
    case actions{2}		# add a frame
      load(fnof);
      mfile = sprintf(mpat, ++frameno);
      drawnow("png", mfile);
      save("-text",fnof,"frameno");
      if (verbose) printf("Frame '%s' added.\n", mfile); endif
    case actions{3}		# close the movie
      switch (mtype)
	case {types{[1 5 7 8 9 10]}} # mp4, ogg, mov, mjpeg, avi, flv
	  cmd = sprintf("ffmpeg -y -r %d -sameq -i %s %s 2>&1", rate, mpat, mmov);
	case {types{[2 3]}}	# mng, gif
	  cmd = sprintf("convert %s -adjoin %s 2>&1", mglob, mpat);
	case types{4}		# zip
	  cmd = sprintf("zip -qr9 %s %s 2>&1", mmov, mglob);
	case types{6}		# swf
	  cmd = sprintf("png2swf -z -r %d -o %s %s", rate, mmov, mglob);
	case types{end}		# dir
	  rename(mdir, mmov); return
	otherwise
	  error("second arg must end with one of:%s", sprintf(" %s",types{:}));
      endswitch
      [status output] = system(cmd);
      if (status != 0)
	load(fnof);
	error("Creation of movie '%s' containing %d frames failed:\n%s",
	      mmov, frameno, output);
      endif
      if (verbose) printf("Movie '%s' contains %d frames:\n%s",
			  mmov, frameno, output); endif
      cleandir(mdir, verbose);
  endswitch
endfunction


function cleandir(mdir, verbose)
  unwind_protect
    save_crr = confirm_recursive_rmdir(false);
    [allgood msg] = rmdir(mdir,"s");
    if (!allgood)
      error("while removing dir '%s': %s", mdir, msg); endif
  unwind_protect_cleanup
    confirm_recursive_rmdir(save_crr);
  end_unwind_protect
  if (verbose) printf("Directory '%s' removed\n", mdir); endif
endfunction

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From jjstickel at vcn.com  Wed Sep 16 08:57:28 2009
From: jjstickel at vcn.com (Jonathan Stickel)
Date: Wed, 16 Sep 2009 07:57:28 -0600
Subject: regdatasmooth
In-Reply-To: <mailman.1701.1253090733.21430.help-octave@octave.org>
References: <mailman.1701.1253090733.21430.help-octave@octave.org>
Message-ID: <4AB0EEC8.3060508@vcn.com>

On 9/16/09 help-octave-request at octave.org wrote:
> Date: Tue, 15 Sep 2009 13:14:25 -0700 (PDT)
> From: esteban1 <estebanvoehringer at hotmail.com>
> Subject: regdatasmooth
> To: help-octave at octave.org
> Message-ID: <25461009.post at talk.nabble.com>
> Content-Type: text/plain; charset=us-ascii
> 
> 
> Hi,
> 
> I have a problem with the regdatasmooth function implemented in the
> data-smoothing-1.2.0 package. Whenever I want to run a demo or the function
> itself I get the following error: 
> error: unary operator `'' not implemented for `list' operands.
> 
> The regdatasmooth function calls the rgdtsmcore function. This function
> works fine, when I run it with my parameters.
> 
> But I would like to use regdatasmooth since it optimizes the smoothing
> through lambda and other parameters.
> I allready searched the web and the mailing list but I was not able to find
> something.
> 
> Thanks for the help!


There have been some changes to the optimization functions in Octave and 
Octave-forge recently.  There is a new fminunc function with a different 
implementation in Octave that takes different arguments.  To get the 
latest regdatasmooth, download it from here:

http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/main/data-smoothing/inst/

Unfortunately, there is a bug in Octave <= 3.2.2 related to cholesky 
factorizations that causes fminunc not to work for single variable 
functions.  A patch is available that may be included in the upcoming 
3.2.3 release (not completely sure).

As a workaround, you may put the fminunc from /Octave-forge/ in a place 
in your path so that it supersedes the fminunc in /Octave/.  If you do 
that, you should use the previous version of regdatasmooth (the one that 
is currently not working for you).

Let me know if you have more trouble or if you get it working.

Jonathan


P.S.
I am moving this thread to octave-dev since it is about an octave-forge 
function.  When replying, please remove "help-octave" from the send list.

From Pascal.Dupuis at worldonline.be  Wed Sep 16 05:22:16 2009
From: Pascal.Dupuis at worldonline.be (Dupuis)
Date: Wed, 16 Sep 2009 03:22:16 -0700 (PDT)
Subject: Randomly generate 10 different integers
In-Reply-To: <7a394d4e0906301045v2c1f13b9gdff231dcac4cdd6@mail.gmail.com>
References: <7a394d4e0906301045v2c1f13b9gdff231dcac4cdd6@mail.gmail.com>
Message-ID: <25469653.post@talk.nabble.com>




Xin Dong wrote:
> 
> Hi,
> 
> I wanna randomly generate 10 different integers in a given range. I know I
> can use random function and discard duplicate values. But is there any
> other
> simple way to do it?
> 
> 

What you're searching for is an implementation of the Fisher-Yates
algortihm, see http://en.wikipedia.org/wiki/Fisher?Yates_shuffle

Here is a possible way:
x=(1:10); n = length(x);
while n >= 2,
 t = ceil(unifrnd(0, n)); %# returns 1 <= t <= n
 if (n != t), dummy = x(n); x(n) = x(t); x(t) = dummy; endif %# permute
 n = n - 1;
endwhile 
-- 
View this message in context: http://www.nabble.com/Randomly-generate-10-different-integers-tp24277099p25469653.html
Sent from the Octave - General mailing list archive at Nabble.com.



From sttt at gmx.de  Wed Sep 16 03:59:40 2009
From: sttt at gmx.de (Steffen Titze)
Date: Wed, 16 Sep 2009 10:59:40 +0200
Subject: Load data with script
In-Reply-To: <19120.38810.727236.986407@segfault.lan>
References: <4AAFCEA7.50903@gmx.de> <19120.38810.727236.986407@segfault.lan>
Message-ID: <20090916085940.226660@gmx.net>


-------- Original-Nachricht --------
> Datum: Wed, 16 Sep 2009 03:45:30 -0400
> Von: "John W. Eaton" <jwe at octave.org>
> An: Steffen Titze <sttt at gmx.de>
> CC: help-octave at octave.org
> Betreff: Load data with script

> On 15-Sep-2009, Steffen Titze wrote:
> 
> | i have written a script to load some values from a data file. With old 
> | version of octave (3.0.3) i don?t had anyproblems. But with the new 
> | version i cann?t load the data file with the script.
> | 
> | if i load the data file with same syntax with out the script, i don?t 
> | have any problems.
> | 
> | x = load ('C:\Test\090811.txt')
> | 
> | Error massage with script
> | error: invalid call to script
> | 
> | How can i solve the problem? I use a windows pc.
> 
> Is there a load.m file somewhere in Octave's path?  Try
> 
>   which load
> 
> and see what it tells you.
> 
> jwe

 "which load" tells me:
 load is a built-in-function

in the Octave's path isn't a load.m file, also on my system i hadn't found a load.m file. 
-- 
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser

From jwe at octave.org  Wed Sep 16 11:09:45 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 16 Sep 2009 12:09:45 -0400
Subject: Load data with script
In-Reply-To: <20090916085940.226660@gmx.net>
References: <4AAFCEA7.50903@gmx.de> <19120.38810.727236.986407@segfault.lan>
	<20090916085940.226660@gmx.net>
Message-ID: <19121.3529.662727.950169@segfault.lan>

On 16-Sep-2009, Steffen Titze wrote:

| -------- Original-Nachricht --------
| > Datum: Wed, 16 Sep 2009 03:45:30 -0400
| > Von: "John W. Eaton" <jwe at octave.org>
| > An: Steffen Titze <sttt at gmx.de>
| > CC: help-octave at octave.org
| > Betreff: Load data with script
| 
| > On 15-Sep-2009, Steffen Titze wrote:
| > 
| > | i have written a script to load some values from a data file. With old 
| > | version of octave (3.0.3) i don?t had anyproblems. But with the new 
| > | version i cann?t load the data file with the script.
| > | 
| > | if i load the data file with same syntax with out the script, i don?t 
| > | have any problems.
| > | 
| > | x = load ('C:\Test\090811.txt')
| > | 
| > | Error massage with script
| > | error: invalid call to script
| > | 
| > | How can i solve the problem? I use a windows pc.
| > 
| > Is there a load.m file somewhere in Octave's path?  Try
| > 
| >   which load
| > 
| > and see what it tells you.
| > 
| > jwe
| 
|  "which load" tells me:
|  load is a built-in-function
| 
| in the Octave's path isn't a load.m file, also on my system i hadn't found a load.m file. 

So what is the script that you are calling, and precisely how are you
calling it?

jwe


From sttt at gmx.de  Wed Sep 16 11:27:37 2009
From: sttt at gmx.de (Steffen Titze)
Date: Wed, 16 Sep 2009 18:27:37 +0200
Subject: Load data with script
In-Reply-To: <19121.3529.662727.950169@segfault.lan>
References: <4AAFCEA7.50903@gmx.de>	<19120.38810.727236.986407@segfault.lan>	<20090916085940.226660@gmx.net>
	<19121.3529.662727.950169@segfault.lan>
Message-ID: <4AB111F9.4050802@gmx.de>

John W. Eaton schrieb:
> On 16-Sep-2009, Steffen Titze wrote:
>
> | -------- Original-Nachricht --------
> | > Datum: Wed, 16 Sep 2009 03:45:30 -0400
> | > Von: "John W. Eaton" <jwe at octave.org>
> | > An: Steffen Titze <sttt at gmx.de>
> | > CC: help-octave at octave.org
> | > Betreff: Load data with script
> | 
> | > On 15-Sep-2009, Steffen Titze wrote:
> | > 
> | > | i have written a script to load some values from a data file. With old 
> | > | version of octave (3.0.3) i don?t had anyproblems. But with the new 
> | > | version i cann?t load the data file with the script.
> | > | 
> | > | if i load the data file with same syntax with out the script, i don?t 
> | > | have any problems.
> | > | 
> | > | x = load ('C:\Test\090811.txt')
> | > | 
> | > | Error massage with script
> | > | error: invalid call to script
> | > | 
> | > | How can i solve the problem? I use a windows pc.
> | > 
> | > Is there a load.m file somewhere in Octave's path?  Try
> | > 
> | >   which load
> | > 
> | > and see what it tells you.
> | > 
> | > jwe
> | 
> |  "which load" tells me:
> |  load is a built-in-function
> | 
> | in the Octave's path isn't a load.m file, also on my system i hadn't found a load.m file. 
>
> So what is the script that you are calling, and precisely how are you
> calling it?
>
> jwe
>
>   
it is a script to load a data file.
The script is only a m file. At the time no function or call to a 
function is in the m file, i open only the data file and transfer the 
value martix to x. I comment the other things out, because at the 
beginning i thought their is a problem with the other code. But with 
verion 3.0.3 i hadn't any problems with the code.

To call the m file i go to the folder of the m file in octave and start 
with "script.m"
The m file and die Data.txt are in the same folder.

Steffen

From jwe at octave.org  Wed Sep 16 12:01:54 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 16 Sep 2009 13:01:54 -0400
Subject: Load data with script
In-Reply-To: <4AB111F9.4050802@gmx.de>
References: <4AAFCEA7.50903@gmx.de> <19120.38810.727236.986407@segfault.lan>
	<20090916085940.226660@gmx.net>
	<19121.3529.662727.950169@segfault.lan> <4AB111F9.4050802@gmx.de>
Message-ID: <19121.6658.414657.793801@segfault.lan>

On 16-Sep-2009, Steffen Titze wrote:

| it is a script to load a data file.
| The script is only a m file. At the time no function or call to a 
| function is in the m file, i open only the data file and transfer the 
| value martix to x. I comment the other things out, because at the 
| beginning i thought their is a problem with the other code. But with 
| verion 3.0.3 i hadn't any problems with the code.
| 
| To call the m file i go to the folder of the m file in octave and start 
| with "script.m"
| The m file and die Data.txt are in the same folder.

So the file is named script.m and you are typing script.m at the
Octave prompt?  Try typing script (without the .m extension).

If that is the problem, then did you never see

  error: can't perform indexing operations for <unknown type> type

when doing this with Octave 3.0.x?

This used to happen because an expression like

  s.e

is an expression to access the structure element E from the structure
S.  When you type

  script.m

Octave evaluated "script" by calling the script then attempting to
index the result with ".m".  So the script was being evaluated, but
since scripts don't return values, the indexing operation failed.

Now, Octave recognizes in advance that "script" is a script, and since
it can't produce a value for indexing, it gives you the error
immediately instead of after evaluating the script.

jwe

From sttt at gmx.de  Wed Sep 16 12:20:23 2009
From: sttt at gmx.de (Steffen Titze)
Date: Wed, 16 Sep 2009 19:20:23 +0200
Subject: Load data with script
In-Reply-To: <19121.6658.414657.793801@segfault.lan>
References: <4AAFCEA7.50903@gmx.de>	<19120.38810.727236.986407@segfault.lan>	<20090916085940.226660@gmx.net>	<19121.3529.662727.950169@segfault.lan>	<4AB111F9.4050802@gmx.de>
	<19121.6658.414657.793801@segfault.lan>
Message-ID: <4AB11E57.7080904@gmx.de>


> So the file is named script.m and you are typing script.m at the
> Octave prompt?  Try typing script (without the .m extension).
>
> If that is the problem, then did you never see
>
>   error: can't perform indexing operations for <unknown type> type
>
> when doing this with Octave 3.0.x?
>
> This used to happen because an expression like
>
>   s.e
>
> is an expression to access the structure element E from the structure
> S.  When you type
>
>   script.m
>
> Octave evaluated "script" by calling the script then attempting to
> index the result with ".m".  So the script was being evaluated, but
> since scripts don't return values, the indexing operation failed.
>
> Now, Octave recognizes in advance that "script" is a script, and since
> it can't produce a value for indexing, it gives you the error
> immediately instead of after evaluating the script.
>
> jwe
>
>   
Thanks it works, when i call the script without the .m extension.
 i think i had never see this error message by Octave 3.0.x,

Tomorrow i will test it on another system, where is Octave 3.0.3 still 
intalled.

Steffen

From dasergatskov at gmail.com  Wed Sep 16 17:32:10 2009
From: dasergatskov at gmail.com (Dmitri A. Sergatskov)
Date: Wed, 16 Sep 2009 17:32:10 -0500
Subject: plotting uint8 data
Message-ID: <892b16670909161532o3c59d446v71d5cedc24d6cfae@mail.gmail.com>

I have problems plotting uint8 type data. The data set seems to get
truncated at index=256. I.e. if I do:

x = uint8(100*(randn(1,300)+1));
plot(x)

then only first 256 data points gets plotted.

This is with 3.2.2 and a recent 4.3 cvs gnuplot....

Does anyone see the same?

Sincerely,

Dmitri.
--

From rob at utk.edu  Wed Sep 16 18:18:32 2009
From: rob at utk.edu (Rob Mahurin)
Date: Wed, 16 Sep 2009 19:18:32 -0400
Subject: plotting uint8 data
In-Reply-To: <892b16670909161532o3c59d446v71d5cedc24d6cfae@mail.gmail.com>
References: <892b16670909161532o3c59d446v71d5cedc24d6cfae@mail.gmail.com>
Message-ID: <20090916231831.GB12933@dyn219245190.dz.ornl.gov>

On Wed, Sep 16, 2009 at 05:32:10PM -0500, Dmitri A. Sergatskov wrote:
> I have problems plotting uint8 type data. The data set seems to get
> truncated at index=256. I.e. if I do:
> 
> x = uint8(100*(randn(1,300)+1));
> plot(x)
> 
> then only first 256 data points gets plotted.
> 
> This is with 3.2.2 and a recent 4.3 cvs gnuplot....
> 
> Does anyone see the same?

I see this with 3.2.3rc4.  A workaround is plot(double(x)).

Cheers,
Rob

From dasergatskov at gmail.com  Wed Sep 16 18:23:10 2009
From: dasergatskov at gmail.com (Dmitri A. Sergatskov)
Date: Wed, 16 Sep 2009 18:23:10 -0500
Subject: plotting uint8 data
In-Reply-To: <20090916231831.GB12933@dyn219245190.dz.ornl.gov>
References: <892b16670909161532o3c59d446v71d5cedc24d6cfae@mail.gmail.com>
	<20090916231831.GB12933@dyn219245190.dz.ornl.gov>
Message-ID: <892b16670909161623q2a676fd5s1559b324f4a8287f@mail.gmail.com>

On Wed, Sep 16, 2009 at 6:18 PM, Rob Mahurin <rob at utk.edu> wrote:
> On Wed, Sep 16, 2009 at 05:32:10PM -0500, Dmitri A. Sergatskov wrote:
>> I have problems plotting uint8 type data. The data set seems to get
>> truncated at index=256. I.e. if I do:
>>
>> x = uint8(100*(randn(1,300)+1));
>> plot(x)
>>
>> then only first 256 data points gets plotted.
>>
>> This is with 3.2.2 and a recent 4.3 cvs gnuplot....
>>
>> Does anyone see the same?
>
> I see this with 3.2.3rc4. ?A workaround is plot(double(x)).

I see it with 3.3.50 as well. I figure out the workaround, but I think
it is a bug.
It is somewhat strange to me that even
plot(1:300, x) would fail.
The data does not get disregarded, it is all plotted at x=256
(this is more obvious if one does plot(x,"o;;") or similar).

>
> Cheers,
> Rob
>

Dmitri.
--


From jwe at octave.org  Wed Sep 16 19:01:35 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 16 Sep 2009 20:01:35 -0400
Subject: plotting uint8 data
In-Reply-To: <892b16670909161623q2a676fd5s1559b324f4a8287f@mail.gmail.com>
References: <892b16670909161532o3c59d446v71d5cedc24d6cfae@mail.gmail.com>
	<20090916231831.GB12933@dyn219245190.dz.ornl.gov>
	<892b16670909161623q2a676fd5s1559b324f4a8287f@mail.gmail.com>
Message-ID: <19121.31839.828690.50008@segfault.lan>

On 16-Sep-2009, Dmitri A. Sergatskov wrote:

| On Wed, Sep 16, 2009 at 6:18 PM, Rob Mahurin <rob at utk.edu> wrote:
| > On Wed, Sep 16, 2009 at 05:32:10PM -0500, Dmitri A. Sergatskov wrote:
| >> I have problems plotting uint8 type data. The data set seems to get
| >> truncated at index=256. I.e. if I do:
| >>
| >> x = uint8(100*(randn(1,300)+1));
| >> plot(x)
| >>
| >> then only first 256 data points gets plotted.
| >>
| >> This is with 3.2.2 and a recent 4.3 cvs gnuplot....
| >>
| >> Does anyone see the same?
| >
| > I see this with 3.2.3rc4. ?A workaround is plot(double(x)).
| 
| I see it with 3.3.50 as well. I figure out the workaround, but I think
| it is a bug.
| It is somewhat strange to me that even
| plot(1:300, x) would fail.
| The data does not get disregarded, it is all plotted at x=256
| (this is more obvious if one does plot(x,"o;;") or similar).

This bug is only present in the gnuplot backend.  With 3.3.50, it
works as you would expect with the fltk backend.

I'm no longer putting any effort into the gnuplot backend, but I would
consider a contributed patch.

jwe


From david.sarraf at yahoo.com  Wed Sep 16 22:00:43 2009
From: david.sarraf at yahoo.com (David Sarraf)
Date: Wed, 16 Sep 2009 20:00:43 -0700 (PDT)
Subject: gnuplot not working in RHEL4 32 bit
In-Reply-To: <25347537.post@talk.nabble.com>
References: <FF554F3FB9EA6143BE2B01509561DFF46A443137B8@EX51.exchserver.com>
	<25347537.post@talk.nabble.com>
Message-ID: <25484388.post@talk.nabble.com>


I had the same problem using Octave under Fedora 11.  Compiling from source
per these instructions produced a working version.  The pre-compiled .RPM
file had numerous other bugs, including crashing with a dump of octave-core
on simple errors such as divide by zero.  The compiled version reports an
error but does not crash the interpreter.

Thanks for the guidance




alynghy wrote:
> 
> The problem is solved on RHEL 5.3 with the same errors :
> 
> 1) Remove gnuplot from redhat :
> 
> yum erase gnuplot
> 
> 2) Remove libraries blas/lapack from redhat :
> 
> yum erase lapack
> yum erase blas
> 
> 3) Remove oldies :
> 
> yum erase compat-*
> yum erase compat-lib*
> 
> 4) Compile the last gnuplot and lapack :
> 
> cd gnuplot-4.2.5
> ./configure
> make
> make install
> cd ..
> 
> AND
> 
> cd lapack-3.2.1
> cp make.inc.example make.inc
> make blaslib
> make all
> cp blas_LINUX.a /usr/local/lib/libblas.a
> cp lapack_LINUX.a /usr/local/lib/liblapack.a
> 
> 5) Compile octave :
> 
> cd octave-3.2.2
> ./configure --with-blas=/usr/local/lib  --with-lapack=/usr/local/lib
> --enable-static LIBS=-lpthread
> make
> make install
> 
> Best regards,
> Benjamin
> 
> 
> 
> kelvin-2 wrote:
>> 
>> I have tried recompiling numerous times but the results have been
>> consistenly unsatisfying.  It should be noted that I am able to
>> successfully install and plot on a 64 bit machine with RHEL4.  Here are
>> the error messages - does anyone know where to look or what might be
>> preventing the tool from working with gnuplot?
>> 
>> using octave-3.2.2 with gnuplot 4.3.0.
>> 
>> bones:octave-3.2.2 >run-octave
>> warning: lo_ieee_init: unrecognized floating point format!
>> 
>> ...
>> 
>> octave:1> plot(1:10)
>> error: get: invalid handle (= 1.79769e+308)
>> error: called from:
>> error:  
>> /z/users/kmccollough/projects/octave-3.2.2/scripts/plot/__plt_get_axis_arg__.m
>> at line 57, column 9
>> error:   /z/users/kmccollough/projects/octave-3.2.2/scripts/plot/plot.m
>> at line 178, column 18
>> octave:1> __gnuplot_version__
>> ans = 4.3.0
>> octave:2>
>> 
>> Thanks,
>> kelvin
>> 
>> 
> 
> 

-- 
View this message in context: http://www.nabble.com/gnuplot-not-working-in-RHEL4-32-bit-tp25224366p25484388.html
Sent from the Octave - General mailing list archive at Nabble.com.


From jwe at octave.org  Wed Sep 16 22:58:50 2009
From: jwe at octave.org (John W. Eaton)
Date: Wed, 16 Sep 2009 23:58:50 -0400
Subject: OOP in Octave
In-Reply-To: <7683DFC4-C36F-408E-A4E7-1A03AD77E352@swissonline.ch>
References: <9830987E-56AA-4947-9FBA-C4C2700A0B1B@swissonline.ch>
	<4AAD26F9.100@dbateman.org>
	<7683DFC4-C36F-408E-A4E7-1A03AD77E352@swissonline.ch>
Message-ID: <19121.46074.541221.808845@segfault.lan>

On 14-Sep-2009, Lukas Reichlin wrote:

| Thanks for your quick reply. Do you know when the new style should be  
| ready to use? (in one year, in two years or Octave 3.4, 3.6 ...)

Octave development is almost entirely done by volunteers.  As of now,
I don't know of anyone who is actively working on this project.

jwe

From highegg at gmail.com  Thu Sep 17 01:08:38 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Thu, 17 Sep 2009 08:08:38 +0200
Subject: Randomly generate 10 different integers
In-Reply-To: <25469653.post@talk.nabble.com>
References: <7a394d4e0906301045v2c1f13b9gdff231dcac4cdd6@mail.gmail.com>
	<25469653.post@talk.nabble.com>
Message-ID: <69d8d540909162308m3a025ef8kea0c6d8827b732cb@mail.gmail.com>

On Wed, Sep 16, 2009 at 12:22 PM, Dupuis <Pascal.Dupuis at worldonline.be> wrote:
>
>
>
> Xin Dong wrote:
>>
>> Hi,
>>
>> I wanna randomly generate 10 different integers in a given range. I know I
>> can use random function and discard duplicate values. But is there any
>> other
>> simple way to do it?
>>
>>
>
> What you're searching for is an implementation of the Fisher-Yates
> algortihm, see http://en.wikipedia.org/wiki/Fisher?Yates_shuffle
>
> Here is a possible way:
> x=(1:10); n = length(x);
> while n >= 2,
> ?t = ceil(unifrnd(0, n)); %# returns 1 <= t <= n
> ?if (n != t), dummy = x(n); x(n) = x(t); x(t) = dummy; endif %# permute
> ?n = n - 1;
> endwhile
> --

Btw., in current tip, randperm is a compiled function and implements
the Fisher-Yates (Knuth) shuffle. It can also produce a set of
permutations at once.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From ct529 at york.ac.uk  Thu Sep 17 11:11:14 2009
From: ct529 at york.ac.uk (Topi, Corrado)
Date: Thu, 17 Sep 2009 17:11:14 +0100
Subject: Solving a very strange minimization problem with Octave
Message-ID: <4AB25FA2.1060507@york.ac.uk>

Dear Octave-ians,

I have to solve the following problem:

min || D_ij - SUM p | SUM k A_pk Q_ij| || (1)

or alternatively

min (SUM_ij (D_ij - SUM p | SUM k A_pk Q_ij|)^2) (2)

where I have to find the values of the Apk>=0 which minimize the 
distance (1) or alternatively the residual sum of square (2).

The problem is the || aroud the second SUM.

Is that feasible with Octave?

All the D_ij, Q_ij are real numbers with

0 <= D_ij <= 1
D_ii =0
D_ij=D_ji

for each i,j belonging to {1 .... n}

I have been breaking my head on it for several days, read tons of 
documentation, and papers, and now I am going to ask as last resort ....

Regards
-- 
Corrado Topi

Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk

From highegg at gmail.com  Thu Sep 17 12:14:32 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Thu, 17 Sep 2009 19:14:32 +0200
Subject: Solving a very strange minimization problem with Octave
In-Reply-To: <4AB25FA2.1060507@york.ac.uk>
References: <4AB25FA2.1060507@york.ac.uk>
Message-ID: <69d8d540909171014xf812ad0g4506922a9868c3c5@mail.gmail.com>

On Thu, Sep 17, 2009 at 6:11 PM, Topi, Corrado <ct529 at york.ac.uk> wrote:
> Dear Octave-ians,
>
> I have to solve the following problem:
>
> min || D_ij - SUM p | SUM k A_pk Q_ij| || (1)
>

given that Q_ij is not dependent on p,k, it is, using your notation,
SUM p | SUM k A_pk Q_ij| = |Q_ij| SUM p | SUM k A_pk| = |Q_ij| * C and
you need to minimize || D_ij - C*|Q_ij|||, i.e. a 2-norm of a linear
combination of two given matrices?

If not, please clarify your expression.


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Thu Sep 17 12:22:31 2009
From: ct529 at york.ac.uk (Topi, Corrado)
Date: Thu, 17 Sep 2009 18:22:31 +0100
Subject: Solving a very strange minimization problem with Octave
In-Reply-To: <69d8d540909171014xf812ad0g4506922a9868c3c5@mail.gmail.com>
References: <4AB25FA2.1060507@york.ac.uk>
	<69d8d540909171014xf812ad0g4506922a9868c3c5@mail.gmail.com>
Message-ID: <4AB27057.6060101@york.ac.uk>

Dear Jaroslav,

I apologize. The expression should read:

min || D_ij - SUM p | SUM k A_pk Q_pkij| || (1)

Regards

Jaroslav Hajek wrote:
> On Thu, Sep 17, 2009 at 6:11 PM, Topi, Corrado <ct529 at york.ac.uk> wrote:
>> Dear Octave-ians,
>>
>> I have to solve the following problem:
>>
>> min || D_ij - SUM p | SUM k A_pk Q_ij| || (1)
>>
> 
> given that Q_ij is not dependent on p,k, it is, using your notation,
> SUM p | SUM k A_pk Q_ij| = |Q_ij| SUM p | SUM k A_pk| = |Q_ij| * C and
> you need to minimize || D_ij - C*|Q_ij|||, i.e. a 2-norm of a linear
> combination of two given matrices?
> 
> If not, please clarify your expression.
> 
> 

-- 
Corrado Topi

Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk

From hinow at uwm.edu  Thu Sep 17 13:54:48 2009
From: hinow at uwm.edu (Peter Hinow)
Date: Thu, 17 Sep 2009 13:54:48 -0500 (CDT)
Subject: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)
Message-ID: <1196695894.8292731253213688010.JavaMail.root@mail02.pantherlink.uwm.edu>

Hello all,

my problem is that compilation goes on for quite a while, but then ends with the following error
g++  -I. -I.. -I../liboctave -I../src -I../libcruft/misc  (.....)
../libcruft/libcruft.so: undefined reference to `r_imag'
(.... many more such messages ...)

If I try
make check
../run-octave --norc --silent --no-history ./fntests.m .
exec: 73: /home/peter/octave-3.2.2/src/octave: not found
and indeed, this directory does not exist. I downloaded the file octave-3.2.2.tar.gz just today from 
http://www.gnu.org/software/octave/download.html

Please let me know what to do or how to diagnose this better.
Thank you and best regards,
       Peter

From hinow at uwm.edu  Thu Sep 17 15:18:26 2009
From: hinow at uwm.edu (Peter Hinow)
Date: Thu, 17 Sep 2009 15:18:26 -0500 (CDT)
Subject: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)
In-Reply-To: <1196695894.8292731253213688010.JavaMail.root@mail02.pantherlink.uwm.edu>
Message-ID: <1130906445.8347091253218706252.JavaMail.root@mail02.pantherlink.uwm.edu>

This problam may have solved itself. It was discussed in a thread
"Compiling 3.2.2 on Ubuntu 9.04" (August 2009)
The recommendation is to run the following two commands

sudo apt-get build-dep octave3.0
sudo apt-get install libftgl-dev



----- Original Message -----
From: "Peter Hinow" <hinow at uwm.edu>
To: help-octave at octave.org
Sent: Thursday, September 17, 2009 1:54:48 PM GMT -06:00 US/Canada Central
Subject: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)

Hello all,

my problem is that compilation goes on for quite a while, but then ends with the following error
g++  -I. -I.. -I../liboctave -I../src -I../libcruft/misc  (.....)
../libcruft/libcruft.so: undefined reference to `r_imag'
(.... many more such messages ...)

If I try
make check
../run-octave --norc --silent --no-history ./fntests.m .
exec: 73: /home/peter/octave-3.2.2/src/octave: not found
and indeed, this directory does not exist. I downloaded the file octave-3.2.2.tar.gz just today from 
http://www.gnu.org/software/octave/download.html

Please let me know what to do or how to diagnose this better.
Thank you and best regards,
       Peter
_______________________________________________
Help-octave mailing list
Help-octave at octave.org
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave

From hinow at uwm.edu  Thu Sep 17 15:42:47 2009
From: hinow at uwm.edu (Peter Hinow)
Date: Thu, 17 Sep 2009 15:42:47 -0500 (CDT)
Subject: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)
In-Reply-To: <489213280.8362791253220118603.JavaMail.root@mail02.pantherlink.uwm.edu>
Message-ID: <882411839.8363241253220167154.JavaMail.root@mail02.pantherlink.uwm.edu>

No, unfortunately not. The problem persists.


----- Original Message -----
From: "Peter Hinow" <hinow at uwm.edu>
To: help-octave at octave.org
Sent: Thursday, September 17, 2009 3:18:26 PM GMT -06:00 US/Canada Central
Subject: Re: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)

This problem may have solved itself. It was discussed in a thread
"Compiling 3.2.2 on Ubuntu 9.04" (August 2009)
The recommendation is to run the following two commands

sudo apt-get build-dep octave3.0
sudo apt-get install libftgl-dev



From storrsjm at email.uc.edu  Thu Sep 17 16:32:37 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Thu, 17 Sep 2009 17:32:37 -0400
Subject: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)
In-Reply-To: <882411839.8363241253220167154.JavaMail.root@mail02.pantherlink.uwm.edu>
References: <489213280.8362791253220118603.JavaMail.root@mail02.pantherlink.uwm.edu>
	<882411839.8363241253220167154.JavaMail.root@mail02.pantherlink.uwm.edu>
Message-ID: <a0daad600909171432o2bafd205u63905422717e289@mail.gmail.com>

Can you post more of the error?

--judd


On Thu, Sep 17, 2009 at 4:42 PM, Peter Hinow <hinow at uwm.edu> wrote:

> No, unfortunately not. The problem persists.
>
>
> ----- Original Message -----
> From: "Peter Hinow" <hinow at uwm.edu>
> To: help-octave at octave.org
> Sent: Thursday, September 17, 2009 3:18:26 PM GMT -06:00 US/Canada Central
> Subject: Re: problem installing Octave 3.2.2 on my desktop (Ubuntu 9.04)
>
> This problem may have solved itself. It was discussed in a thread
> "Compiling 3.2.2 on Ubuntu 9.04" (August 2009)
> The recommendation is to run the following two commands
>
> sudo apt-get build-dep octave3.0
> sudo apt-get install libftgl-dev
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090917/fcae63e0/attachment.html 

From jos at ccrma.stanford.edu  Thu Sep 17 16:47:03 2009
From: jos at ccrma.stanford.edu (Julius Smith)
Date: Thu, 17 Sep 2009 15:47:03 -0600
Subject: freqz question (clarified)
In-Reply-To: <8A83011125D82746BD74D09F439F4E2016AEE15F@oc-exchange-3.kmo b.com>
References: <8A83011125D82746BD74D09F439F4E2016AEE15F@oc-exchange-3.kmob.com>
Message-ID: <200909172147.n8HLlHwm028391@mx3.cae.wisc.edu>

I agree that freqz should simply plot the magnitude and phase 
response of the filter.  Since the arguments to freqz are merely 
filter coefficients B and A, there is no guarantee that the concepts 
of "pass band" or "stop band" make any sense for the given filter. 
For example, what if B/A is an allpass filter?

We ARE guaranteed, on the other hand, that every filter B/A (with 
A(1)==1) has a frequency response, and the most commonly desired 
frequency-response display, in my experience, is magnitude and phase 
versus frequency.

I would be happy implement this change (magnitude in dB, phase in 
degrees), but freqz appears to be in the base Octave distribution, 
while I can only change things in octave-forge, as far as I know.

- jos

At 02:03 PM 9/12/2009, Scott.Raevsky wrote:
>Content-class: urn:content-classes:message
>Content-Type: multipart/alternative;
>         boundary="----_=_NextPart_001_01CA33E4.15CEB18C"
>
>Hello,
>
>Just to clarify my previous question, Matlab's output of freqz is as follows:
>
>freqz(b,a,...) plots the magnitude and unwrapped phase of the 
>frequency response of the filter. The plot is displayed in the 
>current figure window.
>
>Why does Octave's freqz not plot magnitude and phase?  Instead it 
>plots 1) pass band, 2) stop band, and 3) phase.  Plotting a separate 
>pass band and stop band doesn't make sense to me.
>
>Thanks
>
>
>MAIL.KMOB.COM made the following 
>annotations---------------------------------------------------------------------PRIVILEGED 
>AND CONFIDENTIAL COMMUNICATION
>
>This electronic transmission, and any documents attached
>hereto, may contain confidential and/or legally privileged
>information. The information is intended only for use by
>the recipient named above. If you have received this
>electronic message in error, please notify the sender and
>delete the electronic message. Any disclosure, copying,
>distribution, or use of the contents of information
>received in error is strictly prohibited.
>---------------------------------------------------------------------
>_______________________________________________
>Help-octave mailing list
>Help-octave at octave.org
>https://www-old.cae.wisc.edu/mailman/listinfo/help-octave


Julius O. Smith III <jos at ccrma.stanford.edu>
Prof. of Music and Assoc. Prof. (by courtesy) of Electrical Engineering
CCRMA, Stanford University
http://ccrma.stanford.edu/~jos/  


From sievers at cita.utoronto.ca  Fri Sep 18 10:41:11 2009
From: sievers at cita.utoronto.ca (Jonathan Sievers)
Date: Fri, 18 Sep 2009 11:41:11 -0400 (EDT)
Subject: disabling opengl in octave 3.2.2
Message-ID: <alpine.LFD.2.00.0909181127450.29306@sealion.cita.utoronto.ca>

Hey folks - I'm trying to compile octave 3.2.2 from source using icc to 
run on a cluster.  OpenGL is causing problems  - there are 
headers/libraries there which show up during configure but break during 
compilation.  Which is not a problem, I don't need graphics, but I can't 
compile.  I've look through configure several times and can't find a flag 
to disable opengl - does such a thing exist, or is there some other simple 
way to turn it off?  Everything works fine under 3.0.5.

BTW, I thought you might enjoy knowing that I now routinely run an MPIed 
octave across three to four thousand cores.  Thanks for the great work!

Jonathan Sievers

From udippel at uniten.edu.my  Fri Sep 18 19:33:13 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sat, 19 Sep 2009 08:33:13 +0800
Subject: Octave 3.2.2 for Ubuntu9.04?
Message-ID: <4AB426C9.4030601@uniten.edu.my>

It seems that 3.2.2 is the current 'standard', and that compilation is 
not that easy.
Is there any package available? Could someone who has it compiling 
properly simply make it into a .deb, please, and make it available?

Uwe

From jordigh at gmail.com  Sat Sep 19 00:08:17 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermos?=)
Date: Sat, 19 Sep 2009 00:08:17 -0500
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB426C9.4030601@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
Message-ID: <9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>

2009/9/18 Uwe Dippel <udippel at uniten.edu.my>:
> It seems that 3.2.2 is the current 'standard', and that compilation is
> not that easy.
> Is there any package available? Could someone who has it compiling
> properly simply make it into a .deb, please, and make it available?

Rapha?l has been hard at work as usual on making Debian packages:

     http://packages.debian.org/sid/octave3.2

Unfortunately, I think these packages got made after the latest Ubuntu
grabbed Debian's unstable snapshot.

You could try mixing Debian and Ubuntu packages, although at least
from the Debian side, this isn't recommended. You might run into
instability this way. This might be excessive paranoia.


From jpswensen at comcast.net  Sat Sep 19 07:28:55 2009
From: jpswensen at comcast.net (John Swensen)
Date: Sat, 19 Sep 2009 08:28:55 -0400
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
Message-ID: <03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>

Or upgrade to Ubuntu 9.10 (Karmik Koala).  I have been doing my  
testing on 9.10 in a virtual machine and it works great.

John Swensen

On Sep 19, 2009, at 1:08 AM, Jordi Guti?rrez Hermos wrote:

> 2009/9/18 Uwe Dippel <udippel at uniten.edu.my>:
>> It seems that 3.2.2 is the current 'standard', and that compilation  
>> is
>> not that easy.
>> Is there any package available? Could someone who has it compiling
>> properly simply make it into a .deb, please, and make it available?
>
> Rapha?l has been hard at work as usual on making Debian packages:
>
>     http://packages.debian.org/sid/octave3.2
>
> Unfortunately, I think these packages got made after the latest Ubuntu
> grabbed Debian's unstable snapshot.
>
> You could try mixing Debian and Ubuntu packages, although at least
> from the Debian side, this isn't recommended. You might run into
> instability this way. This might be excessive paranoia.
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave



From laura-damon at worldnet.att.net  Sat Sep 19 07:00:41 2009
From: laura-damon at worldnet.att.net (damon christenbury)
Date: Sat, 19 Sep 2009 08:00:41 -0400
Subject: New Octave User
Message-ID: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>

I appear to have installed it correctly, but the graphics will not  
display.  I can issue figure and plot commands, and do not receive an  
error, but I do not see any figures.  I can even issue get(gcf)  
command and get a complete list of the properties, but I cannot see it.
Note also that icons in the applications folder have the 'NO' icon on  
top of the application icon, the one with the circle with the cross  
from 11am to 5 pm.

any ideas?

From udippel at uniten.edu.my  Sat Sep 19 09:34:05 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sat, 19 Sep 2009 22:34:05 +0800
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
Message-ID: <4AB4EBDD.8010206@uniten.edu.my>

Jordi Guti?rrez Hermos wrote:
> Rapha?l has been hard at work as usual on making Debian packages:
>
>      http://packages.debian.org/sid/octave3.2
>
> Unfortunately, I think these packages got made after the latest Ubuntu
> grabbed Debian's unstable snapshot.
>   

Obvious:
octave3.2 depends on libamd2.2.0 (>= 1:3.4.0); however:
  Package libamd2.2.0 is not installed.
 octave3.2 depends on libarpack2 (>= 2.1+parpack96.dfsg); however:
  Package libarpack2 is not installed.
 octave3.2 depends on libcamd2.2.0 (>= 1:3.4.0); however:
  Package libcamd2.2.0 is not installed.
 octave3.2 depends on libccolamd2.7.1 (>= 1:3.4.0); however:
  Package libccolamd2.7.1 is not installed.
 octave3.2 depends on libcholmod1.7.1 (>= 1:3.4.0); however:
  Package libcholmod1.7.1 is not installed.
 octave3.2 depends on libcolamd2.7.1 (>= 1:3.4.0); however:
  Package libcolamd2.7.1 is not installed.
 octave3.2 depends on libcxsparse2.2.3 (>= 1:3.4.0); however:
  Package libcxsparse2.2.3 is not installed.
 octave3.2 depends on libfltk1.1 (>= 1.1.6); however:
  Package libfltk1.1 is not installed.
 octave3.2 depends on libftgl2 (>= 2.1.3~rc5); however:
  Package libftgl2 is not installed.
 octave3.2 depends on libglpk0 (>= 4.30); however:
  Version of libglpk0 on system is 4.29-2.
 octave3.2 depends on libgraphicsmagick++3; however:
  Package libgraphicsmagick++3 is not installed.
 octave3.2 depends on libgraphicsmagick3 (>= 1.3.5); however:
  Package libgraphicsmagick3 is not installed.
 octave3.2 depends on libhdf5-serial-1.8.3 | libhdf5-1.8.3; however:
  Package libhdf5-serial-1.8.3 is not installed.
  Package libhdf5-1.8.3 is not installed.
 octave3.2 depends on libqrupdate1 (>= 1.0); however:
  Package libqrupdate1 is not installed.
 octave3.2 depends on libumfpack5.4.0 (>= 1:3.4.0); however:
  Package libumfpack5.4.0 is not installed.
 octave3.2 depends on octave3.2-common (= 3.2.2-2); however:
  Package octave3.2-common is not installed.
dpkg: error processing octave3.2 (--install):
 dependency problems - leaving unconfigured

but:

$ dpkg -l | grep octave
ii  octave2.9                                  
1:3.0.1-6ubuntu2                          GNU Octave language for 
numerical computations (dummy p
ii  octave3.0                                  
1:3.0.1-6ubuntu2                          GNU Octave language for 
numerical computations (3.0 bra
ii  octave3.0-htmldoc                          
1:3.0.1-6ubuntu2                          HTML documentation on the GNU 
Octave language (3.0 bran

So there are huge changes, after all, I guess.

Thanks anyway,

Uwe


From udippel at uniten.edu.my  Sat Sep 19 09:36:39 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sat, 19 Sep 2009 22:36:39 +0800
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>
Message-ID: <4AB4EC77.7030602@uniten.edu.my>

John Swensen wrote:
> Or upgrade to Ubuntu 9.10 (Karmik Koala).  I have been doing my  
> testing on 9.10 in a virtual machine and it works great.
>   

Yes. But KK is a good month in the future, and I need it (octave) earlier.
So, where do you get KK, and how do you get octave into it??

Uwe


From jpswensen at comcast.net  Sat Sep 19 09:41:48 2009
From: jpswensen at comcast.net (John Swensen)
Date: Sat, 19 Sep 2009 10:41:48 -0400
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB4EC77.7030602@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>
	<4AB4EC77.7030602@uniten.edu.my>
Message-ID: <73C230EE-DD3D-4FDC-9CED-BFA8396250E0@comcast.net>


On Sep 19, 2009, at 10:36 AM, Uwe Dippel wrote:

> John Swensen wrote:
>> Or upgrade to Ubuntu 9.10 (Karmik Koala).  I have been doing my   
>> testing on 9.10 in a virtual machine and it works great.
>>
>
> Yes. But KK is a good month in the future, and I need it (octave)  
> earlier.
> So, where do you get KK, and how do you get octave into it??
>
> Uwe
>

You can upgrade from 9.04 to 9.10 by either downloading the ISO and  
reinstalling completely, or upgrading using the update manager by  
doing the following (found on an Ubuntu mailing list http://ubuntuforums.org/showthread.php?t=1185036 
  and I have used 2 time with no problems):

1) Open a terminal
2) Run update-manager -d
3) Click the button that offers an upgrade.

Hope this helps you get a recent version of Octave up and running more  
quickly.

John Swensen



From joseph.slater at wright.edu  Sat Sep 19 10:22:52 2009
From: joseph.slater at wright.edu (Joseph C. Slater PE, PhD)
Date: Sat, 19 Sep 2009 11:22:52 -0400
Subject: Print font size MacOS X (octave 3.2.2)
Message-ID: <22B9626E-FA1C-4CB9-841D-68C43FFA656F@wright.edu>

The print option
-F:fontsize

where I presume fontsize is a number isn't working on my machine. It  
seems to be adjusting the location of the numbers to account for the  
expected size, but the font size itself is not changing.

Is this just my machine, this version, platform?
Thanks,
Joe

From bpabbott at mac.com  Sat Sep 19 10:35:28 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Sat, 19 Sep 2009 11:35:28 -0400
Subject: New Octave User
In-Reply-To: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>
References: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>
Message-ID: <AA7C7E65-3164-4CF3-B5F3-7F35A24D5696@mac.com>


On Sep 19, 2009, at 8:00 AM, damon christenbury wrote:

> I appear to have installed it correctly, but the graphics will not
> display.  I can issue figure and plot commands, and do not receive an
> error, but I do not see any figures.  I can even issue get(gcf)
> command and get a complete list of the properties, but I cannot see  
> it.
> Note also that icons in the applications folder have the 'NO' icon on
> top of the application icon, the one with the circle with the cross
> from 11am to 5 pm.
>
> any ideas?

Please tell use what version of octave you have installed and what  
operating system you running.

Ben


From bpabbott at mac.com  Sat Sep 19 10:36:50 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Sat, 19 Sep 2009 11:36:50 -0400
Subject: Print font size MacOS X (octave 3.2.2)
In-Reply-To: <22B9626E-FA1C-4CB9-841D-68C43FFA656F@wright.edu>
References: <22B9626E-FA1C-4CB9-841D-68C43FFA656F@wright.edu>
Message-ID: <B27864B3-10FB-4584-A1CA-2078624BA9F6@mac.com>


On Sep 19, 2009, at 11:22 AM, Joseph C. Slater PE, PhD wrote:

> The print option
> -F:fontsize
>
> where I presume fontsize is a number isn't working on my machine. It
> seems to be adjusting the location of the numbers to account for the
> expected size, but the font size itself is not changing.
>
> Is this just my machine, this version, platform?
> Thanks,
> Joe

Joe,

Please provide a simple example so that we may duplicate it.

Ben



From hdfrango at gmail.com  Sat Sep 19 10:24:53 2009
From: hdfrango at gmail.com (hdfrango)
Date: Sat, 19 Sep 2009 08:24:53 -0700 (PDT)
Subject: Some Octave newbie questions regarding number format and matrices
Message-ID: <25522543.post@talk.nabble.com>


Hi,

I started using Octave very recently and am still unsure about a few things:

1. is there any way to force octave to express numbers in fraction form, to
avoid loss of precision?

2. when I want to computer the inverse of a matrix, I have to do inv(A'),
computing the inverse of the transpose, in order to get the correct answer.
Why is that?

thank you
-- 
View this message in context: http://www.nabble.com/Some-Octave-newbie-questions-regarding-number-format-and-matrices-tp25522543p25522543.html
Sent from the Octave - General mailing list archive at Nabble.com.


From storrsjm at email.uc.edu  Sat Sep 19 12:47:12 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 13:47:12 -0400
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB4EC77.7030602@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>
	<4AB4EC77.7030602@uniten.edu.my>
Message-ID: <a0daad600909191047s13be8c88k6890f1fa3fd671f0@mail.gmail.com>

I really don't think it's that hard to compile from sources on Ubuntu. I've
just started a fresh install of 9.04 into a VM and I will write a howto and
post it to the list later today.

--judd



On Sat, Sep 19, 2009 at 10:36 AM, Uwe Dippel <udippel at uniten.edu.my> wrote:

> John Swensen wrote:
> > Or upgrade to Ubuntu 9.10 (Karmik Koala).  I have been doing my
> > testing on 9.10 in a virtual machine and it works great.
> >
>
> Yes. But KK is a good month in the future, and I need it (octave) earlier.
> So, where do you get KK, and how do you get octave into it??
>
> Uwe
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/f88b4ad3/attachment.html 

From dastew at sympatico.ca  Sat Sep 19 13:43:04 2009
From: dastew at sympatico.ca (dastew at sympatico.ca)
Date: Sat, 19 Sep 2009 18:43:04 +0000
Subject: Some Octave newbie questions regarding number format and matrices
In-Reply-To: <25522543.post@talk.nabble.com>
References: <25522543.post@talk.nabble.com>
Message-ID: <COL107-W50CDAFBD4234C1EBBFCCC1A1DF0@phx.gbl>




> Date: Sat, 19 Sep 2009 08:24:53 -0700
> From: hdfrango at gmail.com
> To: help-octave at octave.org
> Subject: Some Octave newbie questions regarding number format and matrices
> 
> 
> Hi,
> 
> I started using Octave very recently and am still unsure about a few things:
> 
> 1. is there any way to force octave to express numbers in fraction form, to
> avoid loss of precision?

I am not sure but I don't think there is.


> 
> 2. when I want to computer the inverse of a matrix, I have to do inv(A'),
> computing the inverse of the transpose, in order to get the correct answer.
> Why is that?
> 

When I do what you say I don't have to do the transpose.: so : What version Of octave do you have and what operating system are you running it on?

example:

octave-3.0.1:1> a= [1 2; 7 3]
a =

   1   2
   7   3

octave-3.0.1:3> b=inv(a)
b =

  -0.272727   0.181818
   0.636364  -0.090909

octave-3.0.1:4> a*b
ans =

   1.00000   0.00000
   0.00000   1.00000

octave-3.0.1:5> 


Doug Stewart



> thank you
> -- 
> View this message in context: http://www.nabble.com/Some-Octave-newbie-questions-regarding-number-format-and-matrices-tp25522543p25522543.html
> Sent from the Octave - General mailing list archive at Nabble.com.
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/e64bd6f3/attachment.html 

From dieter.jurzitza at t-online.de  Sat Sep 19 13:45:44 2009
From: dieter.jurzitza at t-online.de (Dr. Ing. Dieter Jurzitza)
Date: Sat, 19 Sep 2009 20:45:44 +0200
Subject: New Octave User
In-Reply-To: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>
References: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>
Message-ID: <200909192045.45044.dieter.jurzitza@t-online.de>

Hi,
what system are you running? linux, MS-Windows? Anyway, for graphics output 
you need to have gnuplot installed in either case.
For windows, see www.gnuplot.info for downloading options, for linux you 
should be able to install it from the repositories of your distribution.
Hope this helps,
take care



Dieter Jurzitza


Am Samstag 19 September 2009 14:00:41 schrieb damon christenbury:
> I appear to have installed it correctly, but the graphics will not
******

-- 
-----------------------------------------------------------

                               |
                                \
                 /\_/\           |
                | ~x~ |/-----\   /
                 \   /-       \_/
  ^^__   _        /  _  ____   /
 <??__ \- \_/     |  |/    |  |
  ||  ||         _| _|    _| _|

if you really want to see the pictures above - use some font
with constant spacing like courier! :-)
-----------------------------------------------------------



From storrsjm at email.uc.edu  Sat Sep 19 15:27:03 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 16:27:03 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
Message-ID: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>

The following describes how I successfully built octave-3.2.2 on a fresh
install of Ubuntu 9.04. After I finished install from the CD, I used
Synaptic to update all Ubuntu packages. I did not change any software
sources from the defaults. Here's how I proceeded:

1) Install library dependencies
-------------------------------

First try:

$ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev
libarpack2-dev

If it succeeds proceed to step 2. Otherwise if it complains that there is
more than one package that satisfies libcurl4-dev use

$ sudo apt-get install libcurl4-gnutls-dev

and then repeat the first command again.

libcurl4-dev is required for step 2, but it is a virtual package satisfied
by two other packages--either libcurl4-gnutls-dev or libcurl4-openssl-dev. I
used libcurl4-gnutls-dev but I assume you could use the openssl one. If the
first command succeeds then you already have one of these.


2) Install packages needed to build octave
------------------------------------------

$ sudo apt-get build-dep octave3.0


3) Download libqrupdate packages from Karmic. (OPTIONAL)
--------------------------------------------------------

Get and install packages for your platform from:

http://packages.ubuntu.com/karmic/libqrupdate-dev
http://packages.ubuntu.com/karmic/libqrupdate1

Install them using gdebi or

$ dpkg -i libqrupdate1_1.0.1-1_i386.deb libqrupdate-dev_1.0.1-1_i386.deb


4) Download, extract, and build octave-3.2.2
----------------------------------------------

$ wget ftp://ftp.octave.org/pub/octave/octave-3.2.2.tar.bz2
$ tar xf octave-3.2.2.tar.bz2
$ cd octave-3.2.2
$ ./configure --prefix=/usr/local
$ make
$ make check
$ sudo make install


Hope this helps.

--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/43e48ec1/attachment.html 

From tmacchant at yahoo.co.jp  Sat Sep 19 17:06:33 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Sun, 20 Sep 2009 07:06:33 +0900 (JST)
Subject: New Octave User
In-Reply-To: <EBF59771-5E14-4E23-A346-BDF3DB45CD17@worldnet.att.net>
Message-ID: <20090919220633.68197.qmail@web3813.mail.bbt.yahoo.co.jp>

Hello

Please specify what version of octave you have installed and your OS (Linux(disribution name ),
MacOSX, winodows (xp, vista, windows 7).

Please specify show the script you have execute.

Please specify which graphic graphic backend you used. gnuplot or fltk? 

Perhaps you are not familiar with ask the question at the ML.  Please give the detailed information to
explain your trouble.

Regards

Tatsuro



--- damon christenbury  wrote:

> I appear to have installed it correctly, but the graphics will not  
> display.  I can issue figure and plot commands, and do not receive an  
> error, but I do not see any figures.  I can even issue get(gcf)  
> command and get a complete list of the properties, but I cannot see it.
> Note also that icons in the applications folder have the 'NO' icon on  
> top of the application icon, the one with the circle with the cross  
> from 11am to 5 pm.
> 
> any ideas?
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From dastew at sympatico.ca  Sat Sep 19 18:31:42 2009
From: dastew at sympatico.ca (dastew at sympatico.ca)
Date: Sat, 19 Sep 2009 23:31:42 +0000
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
Message-ID: <COL107-W576EFB42E023DF37C1E212A1DF0@phx.gbl>




Date: Sat, 19 Sep 2009 16:27:03 -0400
Subject: Re: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
From: storrsjm at email.uc.edu
To: udippel at uniten.edu.my
CC: Help-octave at octave.org

The following describes how I successfully built octave-3.2.2 on a fresh install of Ubuntu 9.04. After I finished install from the CD, I used Synaptic to update all Ubuntu packages. I did not change any software sources from the defaults. Here's how I proceeded:


1) Install library dependencies
-------------------------------

First try:

$ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev libarpack2-dev

If it succeeds proceed to step 2. Otherwise if it complains that there is more than one package that satisfies libcurl4-dev use


$ sudo apt-get install libcurl4-gnutls-dev

and then repeat the first command again.

libcurl4-dev is required for step 2, but it is a virtual package satisfied by two other packages--either libcurl4-gnutls-dev or libcurl4-openssl-dev. I used libcurl4-gnutls-dev but I assume you could use the openssl one. If the first command succeeds then you already have one of these.



2) Install packages needed to build octave
------------------------------------------

$ sudo apt-get build-dep octave3.0


3) Download libqrupdate packages from Karmic. (OPTIONAL)
--------------------------------------------------------


Get and install packages for your platform from:

http://packages.ubuntu.com/karmic/libqrupdate-dev
http://packages.ubuntu.com/karmic/libqrupdate1


Install them using gdebi or

$ dpkg -i libqrupdate1_1.0.1-1_i386.deb libqrupdate-dev_1.0.1-1_i386.deb


4) Download, extract, and build octave-3.2.2
----------------------------------------------


$ wget ftp://ftp.octave.org/pub/octave/octave-3.2.2.tar.bz2
$ tar xf octave-3.2.2.tar.bz2
$ cd octave-3.2.2
$ ./configure --prefix=/usr/local

$ make
$ make check
$ sudo make install


Hope this helps.

--judd

I tried this and it worked!!

Now How do you add all the packages from source forge?

Doug


 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/c772eab0/attachment.html 

From udippel at uniten.edu.my  Sat Sep 19 21:17:19 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sun, 20 Sep 2009 10:17:19 +0800
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
Message-ID: <4AB590AF.6080403@uniten.edu.my>

Judd Storrs wrote:
> I really don't think it's that hard to compile from sources on Ubuntu.

> The following describes how I successfully built octave-3.2.2 on a 
> fresh install of Ubuntu 9.04. After I finished install from the CD, I 
> used Synaptic to update all Ubuntu packages. I did not change any 
> software sources from the defaults. Here's how I proceeded:
>
> [...]
>
> Hope this helps.

I hope so, too!

Actually, I'm a FOSS-person myself; a Linux user for 10 years. No, 
difficult to do it is not. The thing that's absolutely difficult is the 
How To! How should, how could, anyone guess which packages are 
necessary; which commands are necessary?? I at least couldn't.

And still, it doesn't go through here:

$ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev 
libarpack2-dev
Package libcurl4-dev is a virtual package provided by:
  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
You should explicitly select one to install.
E: Package libcurl4-dev has no installation candidate
$ sudo apt-get install libcurl4-gnutls-dev
Reading package lists... Done
Building dependency tree      
Reading state information... Done
The following extra packages will be installed:
  comerr-dev libgcrypt11-dev libgnutls-dev libgpg-error-dev libidn11-dev
  libkadm55 libkrb5-dev libldap2-dev libtasn1-3-dev
Suggested packages:
  libcurl3-dbg libgcrypt11-doc gnutls-doc gnutls-bin guile-gnutls krb5-doc
The following NEW packages will be installed:
  comerr-dev libcurl4-gnutls-dev libgcrypt11-dev libgnutls-dev
  libgpg-error-dev libidn11-dev libkadm55 libkrb5-dev libldap2-dev
  libtasn1-3-dev
0 upgraded, 10 newly installed, 0 to remove and 0 not upgraded.
Need to get 3928kB of archives.
After this operation, 12.2MB of additional disk space will be used.
Do you want to continue [Y/n]? y
$ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev 
libarpack2-dev
Reading package lists... Done
Building dependency tree      
Reading state information... Done
Package libcurl4-dev is a virtual package provided by:
  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
You should explicitly select one to install.
E: Package libcurl4-dev has no installation candidate
$

And that's the end already. No libcurl4-dev.

So I tried the other way:
$ sudo apt-get install libcurl4-openssl-dev
Reading package lists... Done
Building dependency tree      
Reading state information... Done
The following packages were automatically installed and are no longer 
required:
  libgpg-error-dev libtasn1-3-dev libgcrypt11-dev libgnutls-dev
Use 'apt-get autoremove' to remove them.
The following extra packages will be installed:
  libssl-dev
Suggested packages:
  libcurl3-dbg
The following packages will be REMOVED:
  libcurl4-gnutls-dev
The following NEW packages will be installed:
  libcurl4-openssl-dev libssl-dev
0 upgraded, 2 newly installed, 1 to remove and 0 not upgraded.
Need to get 3035kB of archives.
After this operation, 7029kB of additional disk space will be used.
Do you want to continue [Y/n]? y

$ sudo apt-get build-dep octave3.0
Reading package lists... Done
Building dependency tree      
Reading state information... Done
Package libcurl4-dev is a virtual package provided by:
  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
You should explicitly select one to install.
E: Package libcurl4-dev has no installation candidate
E: Failed to satisfy Build-Depends dependency for octave3.0: libcurl4-dev

Out of luck again, and the end of the line for me.

Thanks anyway,

Uwe



From storrsjm at email.uc.edu  Sat Sep 19 21:51:12 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 22:51:12 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <COL107-W576EFB42E023DF37C1E212A1DF0@phx.gbl>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<COL107-W576EFB42E023DF37C1E212A1DF0@phx.gbl>
Message-ID: <a0daad600909191951j20f7be46x6892413de49b36c9@mail.gmail.com>

I've been pretty piecemeal in my use of octave-forge and I don't remember
running into any problems by just following the instructions in the orange
box on http://octave.sourceforge.net

Basically, I've always just downloaded the desired package.tar.gz from
sourceforge and then I run octave inside the download directory and type:

pkg install package.tar.gz

And everything has magically worked for me. I think that if you've already
built octave you should have everything that is needed (unless the package
has additional requirements--none of the ones I use do).

I don't know if there's a way to build and install them all or to
automatically download dependencies but I think I remember seeing something
about the mentioned in the octave-forge website.

--judd



On Sat, Sep 19, 2009 at 7:31 PM, <dastew at sympatico.ca> wrote:

>
>
> ------------------------------
> Date: Sat, 19 Sep 2009 16:27:03 -0400
> Subject: Re: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
> From: storrsjm at email.uc.edu
> To: udippel at uniten.edu.my
> CC: Help-octave at octave.org
>
>
> The following describes how I successfully built octave-3.2.2 on a fresh
> install of Ubuntu 9.04. After I finished install from the CD, I used
> Synaptic to update all Ubuntu packages. I did not change any software
> sources from the defaults. Here's how I proceeded:
>
> 1) Install library dependencies
> -------------------------------
>
> First try:
>
> $ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev
> libarpack2-dev
>
> If it succeeds proceed to step 2. Otherwise if it complains that there is
> more than one package that satisfies libcurl4-dev use
>
> $ sudo apt-get install libcurl4-gnutls-dev
>
> and then repeat the first command again.
>
> libcurl4-dev is required for step 2, but it is a virtual package satisfied
> by two other packages--either libcurl4-gnutls-dev or libcurl4-openssl-dev. I
> used libcurl4-gnutls-dev but I assume you could use the openssl one. If the
> first command succeeds then you already have one of these.
>
>
> 2) Install packages needed to build octave
> ------------------------------------------
>
> $ sudo apt-get build-dep octave3.0
>
>
> 3) Download libqrupdate packages from Karmic. (OPTIONAL)
> --------------------------------------------------------
>
> Get and install packages for your platform from:
>
> http://packages.ubuntu.com/karmic/libqrupdate-dev
> http://packages.ubuntu.com/karmic/libqrupdate1
>
> Install them using gdebi or
>
> $ dpkg -i libqrupdate1_1.0.1-1_i386.deb libqrupdate-dev_1.0.1-1_i386.deb
>
>
> 4) Download, extract, and build octave-3.2.2
> ----------------------------------------------
>
> $ wget ftp://ftp.octave.org/pub/octave/octave-3.2.2.tar.bz2
> $ tar xf octave-3.2.2.tar.bz2
> $ cd octave-3.2.2
> $ ./configure --prefix=/usr/local
> $ make
> $ make check
> $ sudo make install
>
>
> Hope this helps.
>
> --judd
>
> I tried this and it worked!!
>
> Now How do you add all the packages from source forge?
>
> Doug
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/a40f5e3e/attachment.html 

From storrsjm at email.uc.edu  Sat Sep 19 21:55:17 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 22:55:17 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB590AF.6080403@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
Message-ID: <a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>

That's pretty weird. Are your packages up-to-date?

sudo apt-get update
sudo apt-get upgrade

--judd


On Sat, Sep 19, 2009 at 10:17 PM, Uwe Dippel <udippel at uniten.edu.my> wrote:

> Judd Storrs wrote:
>
>> I really don't think it's that hard to compile from sources on Ubuntu.
>>
>
>  The following describes how I successfully built octave-3.2.2 on a
>> fresh install of Ubuntu 9.04. After I finished install from the CD, I
>> used Synaptic to update all Ubuntu packages. I did not change any
>> software sources from the defaults. Here's how I proceeded:
>>
>> [...]
>>
>> Hope this helps.
>>
>
> I hope so, too!
>
> Actually, I'm a FOSS-person myself; a Linux user for 10 years. No,
> difficult to do it is not. The thing that's absolutely difficult is the
> How To! How should, how could, anyone guess which packages are
> necessary; which commands are necessary?? I at least couldn't.
>
> And still, it doesn't go through here:
>
> $ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev
> libarpack2-dev
> Package libcurl4-dev is a virtual package provided by:
>  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
> You should explicitly select one to install.
> E: Package libcurl4-dev has no installation candidate
> $ sudo apt-get install libcurl4-gnutls-dev
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> The following extra packages will be installed:
>  comerr-dev libgcrypt11-dev libgnutls-dev libgpg-error-dev libidn11-dev
>  libkadm55 libkrb5-dev libldap2-dev libtasn1-3-dev
> Suggested packages:
>  libcurl3-dbg libgcrypt11-doc gnutls-doc gnutls-bin guile-gnutls krb5-doc
> The following NEW packages will be installed:
>  comerr-dev libcurl4-gnutls-dev libgcrypt11-dev libgnutls-dev
>  libgpg-error-dev libidn11-dev libkadm55 libkrb5-dev libldap2-dev
>  libtasn1-3-dev
> 0 upgraded, 10 newly installed, 0 to remove and 0 not upgraded.
> Need to get 3928kB of archives.
> After this operation, 12.2MB of additional disk space will be used.
> Do you want to continue [Y/n]? y
> $ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev
> libarpack2-dev
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Package libcurl4-dev is a virtual package provided by:
>  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
> You should explicitly select one to install.
> E: Package libcurl4-dev has no installation candidate
> $
>
> And that's the end already. No libcurl4-dev.
>
> So I tried the other way:
> $ sudo apt-get install libcurl4-openssl-dev
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> The following packages were automatically installed and are no longer
> required:
>  libgpg-error-dev libtasn1-3-dev libgcrypt11-dev libgnutls-dev
> Use 'apt-get autoremove' to remove them.
> The following extra packages will be installed:
>  libssl-dev
> Suggested packages:
>  libcurl3-dbg
> The following packages will be REMOVED:
>  libcurl4-gnutls-dev
> The following NEW packages will be installed:
>  libcurl4-openssl-dev libssl-dev
> 0 upgraded, 2 newly installed, 1 to remove and 0 not upgraded.
> Need to get 3035kB of archives.
> After this operation, 7029kB of additional disk space will be used.
> Do you want to continue [Y/n]? y
>
> $ sudo apt-get build-dep octave3.0
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Package libcurl4-dev is a virtual package provided by:
>  libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
> You should explicitly select one to install.
> E: Package libcurl4-dev has no installation candidate
> E: Failed to satisfy Build-Depends dependency for octave3.0: libcurl4-dev
>
> Out of luck again, and the end of the line for me.
>
> Thanks anyway,
>
> Uwe
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/a6a110a4/attachment.html 

From udippel at uniten.edu.my  Sat Sep 19 22:09:28 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sun, 20 Sep 2009 11:09:28 +0800
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>	
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
Message-ID: <4AB59CE8.20101@uniten.edu.my>

Judd Storrs wrote:
> That's pretty weird. Are your packages up-to-date?
>
> sudo apt-get update
> sudo apt-get upgrade

Of course! Daily! And without any errors!

Uwe



From udippel at uniten.edu.my  Sat Sep 19 22:15:22 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sun, 20 Sep 2009 11:15:22 +0800
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB59CE8.20101@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>	
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
	<4AB59CE8.20101@uniten.edu.my>
Message-ID: <4AB59E4A.7020204@uniten.edu.my>

Uwe Dippel wrote:
> Judd Storrs wrote:
>> That's pretty weird. Are your packages up-to-date?
>>
>> sudo apt-get update
>> sudo apt-get upgrade
>
> Of course! Daily! And without any errors!

Oh, do we eventually run into an amd64-versus-i386 problem here?
$ uname -a
Linux mybox 2.6.28-15-generic #49-Ubuntu SMP Tue Aug 18 19:25:34 UTC 
2009 x86_64 GNU/Linux


Uwe


From storrsjm at email.uc.edu  Sat Sep 19 22:33:24 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 23:33:24 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB59CE8.20101@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
	<4AB59CE8.20101@uniten.edu.my>
Message-ID: <a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>

What's your output for:

apt-cache showpkg libcurl4-gnutls-dev

and

apt-cache showpkg libcurl4-dev

--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/a9dfca09/attachment.html 

From jordigh at gmail.com  Sat Sep 19 22:44:28 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermos?=)
Date: Sat, 19 Sep 2009 22:44:28 -0500
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB4EBDD.8010206@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<4AB4EBDD.8010206@uniten.edu.my>
Message-ID: <9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>

2009/9/19 Uwe Dippel <udippel at uniten.edu.my>:
> Jordi Guti?rrez Hermos wrote:
>>
>> Rapha?l has been hard at work as usual on making Debian packages:
>>
>> ? ? http://packages.debian.org/sid/octave3.2
>>
>> Unfortunately, I think these packages got made after the latest Ubuntu
>> grabbed Debian's unstable snapshot.
>>
>
> Obvious:
> octave3.2 depends on libamd2.2.0 (>= 1:3.4.0); however:
> ?Package libamd2.2.0 is not installed.

If you want to mix Debian and Ubuntu packages, despite warnings of
possible instability, use apt. That is, add a source line for Debian
repositories in /etc/apt/sources.list and use apt to install them.

The instability is much less likely if the Ubuntu packages are in
Universe, as those are the Debian packages that Ubuntu modifies the
least. I think for the case of Octave, most of the packages are
identical in Debian and Ubuntu.


From storrsjm at email.uc.edu  Sat Sep 19 22:45:38 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sat, 19 Sep 2009 23:45:38 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB59E4A.7020204@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
	<4AB59CE8.20101@uniten.edu.my> <4AB59E4A.7020204@uniten.edu.my>
Message-ID: <a0daad600909192045x26a2f428k3ad3a7e96fa327e3@mail.gmail.com>

On Sat, Sep 19, 2009 at 11:15 PM, Uwe Dippel <udippel at uniten.edu.my> wrote:

> Uwe Dippel wrote:
>
>> Judd Storrs wrote:
>>
>>> That's pretty weird. Are your packages up-to-date?
>>>
>>> sudo apt-get update
>>> sudo apt-get upgrade
>>>
>>
>> Of course! Daily! And without any errors!
>>
>
> Oh, do we eventually run into an amd64-versus-i386 problem here?
> $ uname -a
> Linux mybox 2.6.28-15-generic #49-Ubuntu SMP Tue Aug 18 19:25:34 UTC
> 2009 x86_64 GNU/Linux
>

It's possible, but I doubt it. I use two different x86_64 machines running
9.04 and on both of them have libcurl4-dev provided by libcurl4-gnutls-dev
(but they were installed in March).

The virtual machine I created today was i386, though.

--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090919/1bb0f686/attachment-0001.html 

From udippel at uniten.edu.my  Sat Sep 19 23:13:40 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sun, 20 Sep 2009 12:13:40 +0800
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>	
	<4AB590AF.6080403@uniten.edu.my>	
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>	
	<4AB59CE8.20101@uniten.edu.my>
	<a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>
Message-ID: <4AB5ABF4.4010201@uniten.edu.my>

Judd Storrs wrote:
> What's your output for:
>
> apt-cache showpkg libcurl4-gnutls-dev

$ apt-cache showpkg libcurl4-gnutls-dev
Package: libcurl4-gnutls-dev
Versions:
7.18.2-8ubuntu4.1 
(/var/lib/apt/lists/ubuntu-ashisuto.ubuntulinux.jp_ubuntu_dists_jaunty-updates_main_binary-amd64_Packages) 
(/var/lib/apt/lists/security.ubuntu.com_ubuntu_dists_jaunty-security_main_binary-amd64_Packages) 
(/var/lib/dpkg/status)
 Description Language:
                 File: 
/var/lib/apt/lists/ubuntu-ashisuto.ubuntulinux.jp_ubuntu_dists_jaunty-updates_main_binary-amd64_Packages
                  MD5: 542a849fd833dd4aff3c38a8504ae2b7

7.18.2-8ubuntu4 
(/var/lib/apt/lists/ubuntu-ashisuto.ubuntulinux.jp_ubuntu_dists_jaunty_main_binary-amd64_Packages)
 Description Language:
                 File: 
/var/lib/apt/lists/ubuntu-ashisuto.ubuntulinux.jp_ubuntu_dists_jaunty_main_binary-amd64_Packages
                  MD5: 542a849fd833dd4aff3c38a8504ae2b7


Reverse Depends:
  tclcurl,libcurl4-gnutls-dev
  paintlib-dev,libcurl4-gnutls-dev
  libwfut-0.2-dev,libcurl4-gnutls-dev
  libopkele-dev,libcurl4-gnutls-dev
  libgmyth-dev,libcurl4-gnutls-dev
  libetpan-dev,libcurl4-gnutls-dev 7.16.4-5
  libcurl-ocaml-dev,libcurl4-gnutls-dev 7.15.0-2
  audacious-plugins-dev,libcurl4-gnutls-dev
  python-pycurl,libcurl4-gnutls-dev
  libtunepimp-dev,libcurl4-gnutls-dev
  libraptor1-dev,libcurl4-gnutls-dev
Dependencies:
7.18.2-8ubuntu4.1 - libcurl3-gnutls (5 7.18.2-8ubuntu4.1) libc6-dev (16 
(null)) libc-dev (0 (null)) libgnutls-dev (0 (null)) zlib1g-dev (0 
(null)) libidn11-dev (0 (null)) libkrb5-dev (16 (null)) hurd (0 (null)) 
libldap2-dev (0 (null)) libcurl3-dbg (0 (null)) libcurl-dev (0 (null)) 
libcurl-dev (0 (null)) libcurl-ssl-dev (0 (null)) libcurl3-gnutls-dev (0 
(null)) libcurl4-dev (0 (null))
7.18.2-8ubuntu4 - libcurl3-gnutls (5 7.18.2-8ubuntu4) libc6-dev (16 
(null)) libc-dev (0 (null)) libgnutls-dev (0 (null)) zlib1g-dev (0 
(null)) libidn11-dev (0 (null)) libkrb5-dev (16 (null)) hurd (0 (null)) 
libldap2-dev (0 (null)) libcurl3-dbg (0 (null)) libcurl-dev (0 (null)) 
libcurl-dev (0 (null)) libcurl-ssl-dev (0 (null)) libcurl3-gnutls-dev (0 
(null)) libcurl4-dev (0 (null))
Provides:
7.18.2-8ubuntu4.1 - libcurl4-dev libcurl3-gnutls-dev libcurl-ssl-dev 
libcurl-dev
7.18.2-8ubuntu4 - libcurl4-dev libcurl3-gnutls-dev libcurl-ssl-dev 
libcurl-dev
Reverse Provides:
$

>
> and
>
> apt-cache showpkg libcurl4-dev

$ apt-cache showpkg libcurl4-dev
Package: libcurl4-dev
Versions:

Reverse Depends:
  libcurl4-gnutls-dev,libcurl4-dev
  libnxml0-dev,libcurl4-dev
  libgdal1-dev,libcurl4-dev
  libcurl4-gnutls-dev,libcurl4-dev
Dependencies:
Provides:
Reverse Provides:
libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
libcurl4-gnutls-dev 7.18.2-8ubuntu4
$


Uwe


From udippel at uniten.edu.my  Sat Sep 19 23:26:16 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Sun, 20 Sep 2009 12:26:16 +0800
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>
References: <4AB426C9.4030601@uniten.edu.my>	
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>	
	<4AB4EBDD.8010206@uniten.edu.my>
	<9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>
Message-ID: <4AB5AEE8.4050505@uniten.edu.my>

Jordi Guti?rrez Hermos wrote:
> If you want to mix Debian and Ubuntu packages, despite warnings of
> possible instability, use apt. That is, add a source line for Debian
> repositories in /etc/apt/sources.list and use apt to install them.
>   

No, I don't 'want to mix' any packages!!
I want and need to get Eidors working, which did the trick as it seems 
until Octave 2.9 (or so). A friend of mine, on Windows, downloaded 
3.2.2, and it failed miserably. Then he suggested I try on Linux, and I 
did, on Ubuntu 9.04, and it also failed, not quite as miserable. Then I 
headed here with the error messages, and was told that 'probably' 3.2.2. 
would not fail. So I have been trying since then to check this out, 
hoping it would not fail, but until now can't find a binary package nor 
install from source.

OMG, I'd say. Most projects with the size of Octave have regular updates 
and repositories of binaries available, the current version has been out 
- if I am not mistaken - for 3 months or more. Why does Octave not offer 
recent packages for widely used distributions? I am sure, it could 
attract quite a number of followers, if one didn't have to dig through 
sourcecode stuff, and compilation?
And then some people wonder, when the year of 'Linux on the desktop' 
will finally be there; respectively, why it isn't yet. :(

And if it compiles, it would be rather simple, to make it into a .deb; 
as far as I remember. Did the same years ago for some software.

<end of rant>

Uwe


From tomdean at speakeasy.org  Sat Sep 19 23:59:53 2009
From: tomdean at speakeasy.org (Thomas D. Dean)
Date: Sat, 19 Sep 2009 21:59:53 -0700
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB5AEE8.4050505@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<4AB4EBDD.8010206@uniten.edu.my>
	<9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>
	<4AB5AEE8.4050505@uniten.edu.my>
Message-ID: <1253422793.4629.74.camel@asus>

On Sun, 2009-09-20 at 12:26 +0800, Uwe Dippel wrote:
I have 

GNU Octave, version 3.2.2 on Ubuntu 9.04.

I installed a previous version of octave, I believe 3.0, with apt-get.
Then, installed build depends.

I built octave 3.2.2.  I may have added some -dev packages in response
to error messages, I don't remember.





From storrsjm at email.uc.edu  Sun Sep 20 00:24:17 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Sun, 20 Sep 2009 01:24:17 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB5ABF4.4010201@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
	<4AB59CE8.20101@uniten.edu.my>
	<a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>
	<4AB5ABF4.4010201@uniten.edu.my>
Message-ID: <a0daad600909192224q74030eb0pa127c51120090d53@mail.gmail.com>

Ok, new approach since the libcurl4-dev virtual packge isn't working for you
maybe if you manually install the build dependencies you can skip the
build-dep step.

I got the octave3.0_3.0.1-6ubuntu2.dsc from ubuntu ("apt-get source
octave3.0"). These are the packages listed as build dependencies:

g++-4.1
gcc-4.1
gfortran
debhelper
autoconf
texinfo
texlive-latex-base
texlive-generic-recommended
libreadline5-dev
libncurses5-dev
gperf
libhdf5-serial-dev
libhdf5-lam-dev
libhdf5-mpich-dev
libblas-dev
liblapack-dev
gnuplot-nox
libfftw3-dev
texi2html
less
dpatch
slice
libpcre3-dev
flex
libglpk-dev
libsuitesparse-dev
gawk
ghostscript
libqhull-dev
desktop-file-utils

Also install:

libcurl4-gnutls-dev
libgraphicsmagick++1-dev
libfltk-dev
libarpack2-dev

Then try continuing from step 3

--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090920/0db2054f/attachment.html 

From jordigh at gmail.com  Sun Sep 20 00:46:45 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermos?=)
Date: Sun, 20 Sep 2009 00:46:45 -0500
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB5AEE8.4050505@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<4AB4EBDD.8010206@uniten.edu.my>
	<9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>
	<4AB5AEE8.4050505@uniten.edu.my>
Message-ID: <9543b3a40909192246h33f49d87v83ba3bd324ca912a@mail.gmail.com>

2009/9/19 Uwe Dippel <udippel at uniten.edu.my>:
> Why does Octave not offer recent packages for widely used distributions?

Because it is largely a mistake when a software project makes its own
packages for distributions instead of helping or working closely with
the distribution to directly incorporate a package for, well,
distribution.

Packaging for a distribution is a considerable task. Each distribution
has slighlty differing rules and policies on how compliation should be
made, which packages should be offered, where files should go, how
documentation should be presented, how should dependencies be
resolved, how source should be offered, etc. etc.

Too often software projects will offer their own .debs that are buggy
in minute ways because they do not adhere to a distribution's
packaging policy. Sure, they're often "good enough" and work, but we
should demand certain minimum

> And then some people wonder, when the year of 'Linux on the desktop' will
> finally be there; respectively, why it isn't yet. :(

I understand your frustration, and it is ours all at least part of the
time, but you know what they say about candles and cursing the
darkness. We are all volunteers.

> And if it compiles, it would be rather simple, to make it into a .deb; as
> far as I remember. Did the same years ago for some software.

It's possible to make a quick-and-dirty .deb that probably works for
you, but it's really nothing more than a convenience of tracking the
default Octave installation through apt. It is highly unlikely that
the default installation of any software package will adhere to Ubuntu
or Debian policies, so Octave itself cannot offer such .debs directly
and must instead rely on specific packagers like Rafael Laboissiere,
Thomas Weber, and infrequently myself.

Given your evident frustration, I think the easiest route would be to
either follow up on what I suggested and install Debian Octave's with
apt and do so carefully, paying close attention to which libraries
will get installed from each source, and use apt pinning (don't
download .deb's manually and expect them to satisfy dependencies)

OR

follow up on John Swenson and upgrade your Ubuntu prematurely to the
next release of Ubuntu which has the latest Octave packages

OR

build Octave from source. After you've downloaded the source, your
biggest problem with compiling it will no doubt be satisfying
dependencies. You can use a little Debian (or Ubuntu) help to satisfy
build dependencies by running "apt-get build-dep octave3.2" to get
everything you should need to build Octave before you run ./configure
&& make && sudo checkinstall make install (apt-get build checkinstall
to download checkinstall which will make your own aforementioned
quick-and-dirty .deb).

HTH,
- Jordi G. H.

From jordigh at gmail.com  Sun Sep 20 01:06:58 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermos?=)
Date: Sun, 20 Sep 2009 01:06:58 -0500
Subject: Some Octave newbie questions regarding number format and matrices
In-Reply-To: <25522543.post@talk.nabble.com>
References: <25522543.post@talk.nabble.com>
Message-ID: <9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>

2009/9/19 hdfrango <hdfrango at gmail.com>:
> 1. is there any way to force octave to express numbers in fraction form, to
> avoid loss of precision?

If you don't want to lose precision, you need a symbolic CAS, like
Sage or Maxima, not Octave. Octave is optimised for speed with
hardware floats, and although it does have a symbolic addon in the
Octave-Forge packages, this is more of an afterthought than core
functionality. In particular, it only allows you to change the
precision. I don't believe a bignum interface exists for Octave.

> 2. when I want to computer the inverse of a matrix, I have to do inv(A'),
> computing the inverse of the transpose, in order to get the correct answer.
> Why is that?

My guess is that you've either found a bizarre bug in Octave, or you
are confused about what the inverse of a matrix should be. Can you
show us a specific matrix for which inv(A) produces the transpose of
the correct inverse?

- Jordi G. H.

From jordigh at gmail.com  Sun Sep 20 01:30:30 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermos?=)
Date: Sun, 20 Sep 2009 01:30:30 -0500
Subject: help me
In-Reply-To: <fbe9db0c0909132343w10556861t51a5859271c59061@mail.gmail.com>
References: <fbe9db0c0909132343w10556861t51a5859271c59061@mail.gmail.com>
Message-ID: <9543b3a40909192330n78db236cm8d1e2976bcb8028f@mail.gmail.com>

2009/9/14 diego vega <linuxdieg at gmail.com>:

> quisiera saber si existen o hay que crearlas; las funciones son:
> funcion escalon, funcion signo, pulso triangular , inpulso unitario.

(in Spanish, Diego is asking if Octave has a step function, a sign
function, unit impulse, and triangular pulse. English answer follows
below Spanish answer.)

Hola Diego. El idioma de la lista es ingl?s, pero varios de nosotros
podemos responder en castellano. A?n as?, por favor escribe en ingl?s,
pues es el idioma que la mayor?a de suscriptores de la lista hablan.

Las funciones que solicitas son muy f?ciles de generar. Por ejemplo,
una funci?n escal?n se puede generar f?cilmente a partir las funciones
piso y techo (en Octave llamadas floor y ceil). La funci?n signo ya
est? implantada y se llama sign en Octave. Creo que con pulso triangular
te refieres a una funci?n parecida a ?sta:

     x.*(heaviside(x) - 2*heaviside(x-1)) + 2*heaviside(x-1)

La funci?n Heaviside se encuentra en el paquete de funciones
especiales (specfun) de Octave-Forge.

No s? qu? es un impulso unitario, pero me imagino que podr?s generarlo
tambi?n con la funci?n de Heaviside.

Espero esto te ayude,
- Jordi G. H.

================== English translation =====================

Hello, Diego. The mailing list language is English, but several of us
can answer in Spanish. Even so, please write in English, as it's the
language that most of the mailing list subscribers can speak.

The functions you request are very easy to generate. For example, a
step function can easily be generated by floor and ceiling functions
(in Octave, they're called floor and ceil). The sign function is
already implemented and it's called sign in Octave. I think by
triangular pulse you mean something like this:

     x.*(heaviside(x) - 2*heaviside(x-1)) + 2*heaviside(x-1)

The Heaviside function is in the special functions package (specfun)
in Octave-Forge.

I don't konw what a unit impulse is, but I can imagine that you can
also generate it with the Heaviside function.

HTH,
- Jordi G. H.


From Potorti at isti.cnr.it  Sun Sep 20 04:04:12 2009
From: Potorti at isti.cnr.it (Francesco Potorti`)
Date: Sun, 20 Sep 2009 11:04:12 +0200
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <4AB5AEE8.4050505@uniten.edu.my>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<4AB4EBDD.8010206@uniten.edu.my>
	<9543b3a40909192044m6c0b7a8etaffc2b8a573ce4eb@mail.gmail.com>
	<4AB5AEE8.4050505@uniten.edu.my>
Message-ID: <E1MpIL6-0002Wl-4Q@tucano.isti.cnr.it>

>And if it compiles, it would be rather simple, to make it into a .deb; 
>as far as I remember. Did the same years ago for some software.

Well, it is into a .deb, but since it is new you have to have a very new
intallation.  What I do, on Debian testing, is to download the
experimental .deb packages by hand and install them.  I suppose that apt
pinning would make my life easier, but this is what I do now and it
works.

Using the nice debtree package at <http://alioth.debian.org/~fjp/debtree/>
I built this graphical tree with
  debtree --max-depth=2 -I octave3.2|dot -Tsvg > octave3.2-deps.svg
which produces a graph, in png and svg, that may be useful for those
who must choose which packages to download:
<http://fly.isti.cnr.it/pub/software/octave/octave3.2-deps.svg>
<http://fly.isti.cnr.it/pub/software/octave/octave3.2-deps.png>

-- 
Francesco Potort? (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/

From thomas.weber.mail at gmail.com  Sun Sep 20 04:18:31 2009
From: thomas.weber.mail at gmail.com (Thomas Weber)
Date: Sun, 20 Sep 2009 11:18:31 +0200
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB590AF.6080403@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
Message-ID: <20090920091831.GB12282@atlan>

On Sun, Sep 20, 2009 at 10:17:19AM +0800, Uwe Dippel wrote:
> Judd Storrs wrote:
> > I really don't think it's that hard to compile from sources on Ubuntu.
> 
> > The following describes how I successfully built octave-3.2.2 on a 
> > fresh install of Ubuntu 9.04. After I finished install from the CD, I 
> > used Synaptic to update all Ubuntu packages. I did not change any 
> > software sources from the defaults. Here's how I proceeded:
> >
> > [...]
> >
> > Hope this helps.
> 
> I hope so, too!
> 
> Actually, I'm a FOSS-person myself; a Linux user for 10 years. No, 
> difficult to do it is not. The thing that's absolutely difficult is the 
> How To! How should, how could, anyone guess which packages are 
> necessary; which commands are necessary?? I at least couldn't.
> 
> And still, it doesn't go through here:
> 
> $ sudo apt-get install libcurl4-dev libgraphicsmagick++1-dev libfltk-dev 
> libarpack2-dev
> Package libcurl4-dev is a virtual package provided by:
>   libcurl4-gnutls-dev 7.18.2-8ubuntu4.1
             ^^^^^^
<snip>
> So I tried the other way:
> $ sudo apt-get install libcurl4-openssl-dev

If apt tells you it has the gnutls package, why do you try the openssl
one?

	Thomas

From hdfrango at gmail.com  Sun Sep 20 06:32:35 2009
From: hdfrango at gmail.com (hdfrango)
Date: Sun, 20 Sep 2009 04:32:35 -0700 (PDT)
Subject: Some Octave newbie questions regarding number format and matrices
In-Reply-To: <9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>
References: <25522543.post@talk.nabble.com>
	<9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>
Message-ID: <25529605.post@talk.nabble.com>



Jordi Guti?rrez Hermos wrote:
> 
> 2009/9/19 hdfrango :
>> 1. is there any way to force octave to express numbers in fraction form,
>> to
>> avoid loss of precision?
> 
> If you don't want to lose precision, you need a symbolic CAS, like
> Sage or Maxima, not Octave. Octave is optimised for speed with
> hardware floats, and although it does have a symbolic addon in the
> Octave-Forge packages, this is more of an afterthought than core
> functionality. In particular, it only allows you to change the
> precision. I don't believe a bignum interface exists for Octave.
> 
>> 2. when I want to computer the inverse of a matrix, I have to do inv(A'),
>> computing the inverse of the transpose, in order to get the correct
>> answer.
>> Why is that?
> 
> My guess is that you've either found a bizarre bug in Octave, or you
> are confused about what the inverse of a matrix should be. Can you
> show us a specific matrix for which inv(A) produces the transpose of
> the correct inverse?
> 
> - Jordi G. H
> 
> 

First of all thank you both for the replies.

1. I will give the tools mentioned a try.

2. As for the matrix issue, I feel really stupid, it turns out even though I
knew how to correctly input a matrix in octave, I was somehow entering the
transpose, which is why I needed to compute the inverse of the transpose in
order to get the expected result :).

thanks again

-- 
View this message in context: http://www.nabble.com/Some-Octave-newbie-questions-regarding-number-format-and-matrices-tp25522543p25529605.html
Sent from the Octave - General mailing list archive at Nabble.com.



From dastew at sympatico.ca  Sun Sep 20 08:54:42 2009
From: dastew at sympatico.ca (dastew at sympatico.ca)
Date: Sun, 20 Sep 2009 13:54:42 +0000
Subject: Some Octave newbie questions regarding number format and matrices
In-Reply-To: <25529605.post@talk.nabble.com>
References: <25522543.post@talk.nabble.com>
	<9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>
Message-ID: <COL107-W20BA34464B147A4401F0ECA1DE0@phx.gbl>




> Date: Sun, 20 Sep 2009 04:32:35 -0700
> From: hdfrango at gmail.com
> To: help-octave at octave.org
> Subject: Re: Some Octave newbie questions regarding number format and matrices
> 
> 
> 
> Jordi Guti?rrez Hermos wrote:
> > 
> > 2009/9/19 hdfrango :
> >> 1. is there any way to force octave to express numbers in fraction form,
> >> to
> >> avoid loss of precision?
> > 
> > If you don't want to lose precision, you need a symbolic CAS, like
> > Sage or Maxima, not Octave. Octave is optimised for speed with
> > hardware floats, and although it does have a symbolic addon in the
> > Octave-Forge packages, this is more of an afterthought than core
> > functionality. In particular, it only allows you to change the
> > precision. I don't believe a bignum interface exists for Octave.
> > 
> >> 2. when I want to computer the inverse of a matrix, I have to do inv(A'),
> >> computing the inverse of the transpose, in order to get the correct
> >> answer.
> >> Why is that?
> > 
> > My guess is that you've either found a bizarre bug in Octave, or you
> > are confused about what the inverse of a matrix should be. Can you
> > show us a specific matrix for which inv(A) produces the transpose of
> > the correct inverse?
> > 
> > - Jordi G. H
> > 
> > 
> 
> First of all thank you both for the replies.
> 
> 1. I will give the tools mentioned a try.
> 
> 2. As for the matrix issue, I feel really stupid, it turns out even though I
> knew how to correctly input a matrix in octave, I was somehow entering the
> transpose, which is why I needed to compute the inverse of the transpose in
> order to get the expected result :).
> 
> thanks again
> 
> -- 
Hey. we all have done something like that in the past so smile.

If you have any more questions , just ask :-)

Doug
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090920/7d83d877/attachment.html 

From m-wright at eskimo.com  Sun Sep 20 17:46:56 2009
From: m-wright at eskimo.com (Maynard Wright)
Date: Sun, 20 Sep 2009 15:46:56 -0700
Subject: help me
In-Reply-To: <9543b3a40909192330n78db236cm8d1e2976bcb8028f@mail.gmail.com>
References: <fbe9db0c0909132343w10556861t51a5859271c59061@mail.gmail.com>
	<9543b3a40909192330n78db236cm8d1e2976bcb8028f@mail.gmail.com>
Message-ID: <200909201546.56749.m-wright@eskimo.com>

On Saturday 19 September 2009 23:30, Jordi Guti?rrez Hermos wrote:
> 2009/9/14 diego vega <linuxdieg at gmail.com>:
> > quisiera saber si existen o hay que crearlas; las funciones son:
> > funcion escalon, funcion signo, pulso triangular , inpulso unitario.
>
> (in Spanish, Diego is asking if Octave has a step function, a sign
> function, unit impulse, and triangular pulse. English answer follows
> below Spanish answer.)
>
> Hola Diego. El idioma de la lista es ingl?s, pero varios de nosotros
> podemos responder en castellano. A?n as?, por favor escribe en ingl?s,
> pues es el idioma que la mayor?a de suscriptores de la lista hablan.
>
> Las funciones que solicitas son muy f?ciles de generar. Por ejemplo,
> una funci?n escal?n se puede generar f?cilmente a partir las funciones
> piso y techo (en Octave llamadas floor y ceil). La funci?n signo ya
> est? implantada y se llama sign en Octave. Creo que con pulso triangular
> te refieres a una funci?n parecida a ?sta:
>
>      x.*(heaviside(x) - 2*heaviside(x-1)) + 2*heaviside(x-1)
>
> La funci?n Heaviside se encuentra en el paquete de funciones
> especiales (specfun) de Octave-Forge.
>
> No s? qu? es un impulso unitario, pero me imagino que podr?s generarlo
> tambi?n con la funci?n de Heaviside.
>
> Espero esto te ayude,
> - Jordi G. H.
>
> ================== English translation =====================
>
> Hello, Diego. The mailing list language is English, but several of us
> can answer in Spanish. Even so, please write in English, as it's the
> language that most of the mailing list subscribers can speak.
>
> The functions you request are very easy to generate. For example, a
> step function can easily be generated by floor and ceiling functions
> (in Octave, they're called floor and ceil). The sign function is
> already implemented and it's called sign in Octave. I think by
> triangular pulse you mean something like this:
>
>      x.*(heaviside(x) - 2*heaviside(x-1)) + 2*heaviside(x-1)
>
> The Heaviside function is in the special functions package (specfun)
> in Octave-Forge.
>
> I don't konw what a unit impulse is, but I can imagine that you can
> also generate it with the Heaviside function.
>
> HTH,
> - Jordi G. H.
>


The unit impulse is also sometimes called the impulse function or the delta 
function.  A non-rigorous definition would be the limit of a rectangular 
pulse (amplitude as a function of time) as the duration in time goes to zero 
and the area is held constant at 1.  Note that the amplitude approaches 
infinity as the duration (width) approaches zero.  The frequency spectrum of 
the unit impulse has a constant amplitude of 1 at all frequencies.


Maynard Wright


From udippel at uniten.edu.my  Sun Sep 20 22:54:34 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Mon, 21 Sep 2009 11:54:34 +0800
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <a0daad600909192224q74030eb0pa127c51120090d53@mail.gmail.com>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>	
	<4AB590AF.6080403@uniten.edu.my>	
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>	
	<4AB59CE8.20101@uniten.edu.my>	
	<a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>	
	<4AB5ABF4.4010201@uniten.edu.my>
	<a0daad600909192224q74030eb0pa127c51120090d53@mail.gmail.com>
Message-ID: <4AB6F8FA.1090108@uniten.edu.my>

Judd Storrs wrote:
> Ok, new approach since the libcurl4-dev virtual packge isn't working 
> for you maybe if you manually install the build dependencies you can 
> skip the build-dep step.
>
> I got the octave3.0_3.0.1-6ubuntu2.dsc from ubuntu ("apt-get source 
> octave3.0"). These are the packages listed as build dependencies:
>
> g++-4.1
> gcc-4.1
> gfortran
> debhelper
> autoconf
> texinfo
> texlive-latex-base
> texlive-generic-recommended
> libreadline5-dev
> libncurses5-dev
> gperf
> libhdf5-serial-dev
> libhdf5-lam-dev
> libhdf5-mpich-dev
> libblas-dev
> liblapack-dev
> gnuplot-nox
> libfftw3-dev
> texi2html
> less
> dpatch
> slice
> libpcre3-dev
> flex
> libglpk-dev
> libsuitesparse-dev
> gawk
> ghostscript
> libqhull-dev
> desktop-file-utils
>
> Also install:
>
> libcurl4-gnutls-dev
> libgraphicsmagick++1-dev
> libfltk-dev
> libarpack2-dev
>
> Then try continuing from step 3

Thanks so much, Judd!

I can confirm that Octave 3.2.2 compiles and installs with these 
dependencies, with some minor modifications, also suggested by you:
leave out: libhdf5-lam-dev libhdf5-mpich-dev
add: libftgl-dev libfltk-dev

and it goes through.

Next, time to try Eidors.
octave:1> run /home/udippel/Sources/eidors-v3.3/startup.m
warning: you do not have a compiled mex file eidors_var_id.
Please compile it using: mex 
/home/udippel/Sources/eidors-v3.3/arch/eidors_var_id.cpp

[exit]

This is pdfTeXk, Version 3.141592-1.40.3 (Web2C 7.5.6)
 %&-line parsing enabled.
entering extended mode
(/home/udippel/Sources/eidors-v3.3/arch/eidors_var_id.cpp
This is MeX  Version 1.05  18 XII 1993  (B. Jackowski & M. Ry\'cko)
! Missing $ inserted.
<inserted text>
                $
<to be read again>
                   _
l.1 /* eidors_
              var.id: calculate the sha1 hash of variable in memory
?

What now?

At least, demo_real.m works.

Thanks a bunch!

Uwe


From udippel at uniten.edu.my  Sun Sep 20 23:09:41 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Mon, 21 Sep 2009 12:09:41 +0800
Subject: Eidors and Octave
Message-ID: <4AB6FC85.5080404@uniten.edu.my>

[continuation of earlier thread, in which 3.2.2. was suggested]

Finally, I got 3.2.2 installed.

Alas, out of the Eidors examples:
cheating_2d.m       demo_3d_simdata.m  manchester_tomography.m
compare_2d_algs.m  demo_complex.m     moving_tank_objs.m
compare_3d_algs.m  demo_real.m          resistor_model.m
demo_2d_simdata.m  image_2d_algs.m  
only demo_2d_simdata.m and demo_real.m work.
The others fail one way or another. About half with
structure has no member `cache_priority'
Though Google has no result for this.
What could I do, please?

Uwe



From storrsjm at email.uc.edu  Sun Sep 20 23:41:01 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Mon, 21 Sep 2009 00:41:01 -0400
Subject: Octave 3.2.2 for Ubuntu9.04 -- How to compile from sources
In-Reply-To: <4AB6F8FA.1090108@uniten.edu.my>
References: <a0daad600909191327h7b93b699g92b6d15c7ed187b8@mail.gmail.com>
	<4AB590AF.6080403@uniten.edu.my>
	<a0daad600909191955t60f63746k7a68996885811c09@mail.gmail.com>
	<4AB59CE8.20101@uniten.edu.my>
	<a0daad600909192033g908a2fer139984184056abac@mail.gmail.com>
	<4AB5ABF4.4010201@uniten.edu.my>
	<a0daad600909192224q74030eb0pa127c51120090d53@mail.gmail.com>
	<4AB6F8FA.1090108@uniten.edu.my>
Message-ID: <a0daad600909202141s67b1a1c0k69c53c86b67c0497@mail.gmail.com>

Ok, I don't know eidors at all, but that error about a missing mex file
looked simple enough. I downloaded eidors and extracted it to look around.
It looks like startup.m expects to be running on octave in windows and
doesn't set the path correctly to the mex file otherwise. Also they don't
provied a linux version of the mex file so I needs to be compiled. I only
found one mex file in there. This gets the startup working:

$ cd eidors-v3.3
$ cd arch
$ mkoctfile --mex eidors_var_id.cpp
$ cp eidors_var_id.mex ..
$ cd ..
$ octave
octave:1> startup
warning: isstr is obsolete and will be removed from a future version of
Octave, please use ischar instead
EIDORS:[Completed setting up of EIDORS Version 3.3 ($Date: 2008/07/19
16:14:18 $)]
EIDORS:[Parameter: cache_size=100 MB]
EIDORS:[Parameter: mapped_colour=127]
EIDORS:[Default background colour: write]
EIDORS:[EIDORS mex folder: /home/judd/Desktop/eidors/eidors-v3.3]


--judd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090921/f4226d90/attachment.html 

From ct529 at york.ac.uk  Mon Sep 21 04:01:50 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 21 Sep 2009 10:01:50 +0100
Subject: Compiling with and without -march=native: benchmarks
Message-ID: <200909211001.50518.ct529@york.ac.uk>

Dear Octave friends,

here is the benchmark comparison between Octave 322 on Kubuntu 904 64 bit 
compiled from source with and without the --march=native flag set.

It seem that with the new gcc the flag -march-native yields higher 
performances. 

What do you think?

PS: I believe recompiling lapack and blas with -march=native may lead to even 
higher performances, but I have found lapack / blas recompilation rather 
troublesome.

Regards 

COMPILED WITH -O2 -MARCH=NATIVE
octave:1> benchmark
Octave benchmark version bm 2.3
Speed of octave 3.2.2 on x86_64-unknown-linux-gnu
  Pentium II 350 MHz has score 1.0, higher numbers are better
Matrix inversion (LAPACK)        14.98 +/- 0.1% (252 runs)
Schur decomposition (LAPACK)     54.52 +/- 0.4% (322 runs)
Differential equation (LSODE)    40.81 +/- 0.2% (980 runs)
Fourier transforms (FFTPACK)     23.09 +/- 0.4% (609 runs)
for loop                         75.20 +/- 3.6% (1365 runs)
-- Performance index (bm 2.3): 36

COMPILED WITH -O2
octave:4> benchmark
Octave benchmark version bm 2.3
Speed of octave 3.2.2 on x86_64-unknown-linux-gnu
  Pentium II 350 MHz has score 1.0, higher numbers are better
Matrix inversion (LAPACK)        15.11 +/- 0.3% (259 runs)
Schur decomposition (LAPACK)     54.36 +/- 0.2% (322 runs)
Differential equation (LSODE)    38.54 +/- 0.2% (910 runs)
Fourier transforms (FFTPACK)     21.55 +/- 2.8% (567 runs)
for loop                         63.70 +/- 1.0% (1183 runs)
-- Performance index (bm 2.3): 34




-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From udippel at uniten.edu.my  Mon Sep 21 04:11:27 2009
From: udippel at uniten.edu.my (Uwe Dippel)
Date: Mon, 21 Sep 2009 17:11:27 +0800
Subject: Octave 3.2.2 for Ubuntu9.04?
In-Reply-To: <73C230EE-DD3D-4FDC-9CED-BFA8396250E0@comcast.net>
References: <4AB426C9.4030601@uniten.edu.my>
	<9543b3a40909182208t777f251bn30050dc2a44d92a7@mail.gmail.com>
	<03CAB1BE-0146-4EA5-AE89-AE46AEC0F187@comcast.net>
	<4AB4EC77.7030602@uniten.edu.my>
	<73C230EE-DD3D-4FDC-9CED-BFA8396250E0@comcast.net>
Message-ID: <4AB7433F.7030001@uniten.edu.my>

John Swensen wrote:
>
> You can upgrade from 9.04 to 9.10 by either downloading the ISO and  
> reinstalling completely, or upgrading using the update manager by  
> doing the following (found on an Ubuntu mailing list http://ubuntuforums.org/showthread.php?t=1185036 
>   and I have used 2 time with no problems):
>
> 1) Open a terminal
> 2) Run update-manager -d
> 3) Click the button that offers an upgrade.
>
> Hope this helps you get a recent version of Octave up and running more  
> quickly.
>   

Don't. Ever.
I'm lucky I didn't do that in the first place, now I *tried* on a 
machine on the side, not my 'production' desktop.

> I upgraded from a working and freshly updated 9.04 (apt-get update && 
> apt-get upgrade) to Alpha 6, with
> update-manager -d
> after it had worked well on my netbook.
> Despite of the upgrade going through without a hitch, after reboot 
> everything is damaged, it seems. I get UNEXPECTED INCONSISTENCIES just 
> flashing by, in circles. And this with the old as well as the new 
> kernel, including recovery mode. I can stop the going round in circles 
> with Ctrl-Alt-Del, which will gracefully shut down the machine.

There are lots of people with a similar problem, usually it wouldn't go 
in circles, but still, not boot at all.
https://bugs.launchpad.net/ubuntu/+bug/433829
https://bugs.launchpad.net/ubuntu/+source/e2fsprogs/+bug/427822
https://bugs.launchpad.net/ubuntu/+source/clock-setup/+bug/423247

Uwe


From highegg at gmail.com  Mon Sep 21 04:21:46 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 21 Sep 2009 11:21:46 +0200
Subject: Compiling with and without -march=native: benchmarks
In-Reply-To: <200909211001.50518.ct529@york.ac.uk>
References: <200909211001.50518.ct529@york.ac.uk>
Message-ID: <69d8d540909210221k6146b027rf404b22dfca1d468@mail.gmail.com>

On Mon, Sep 21, 2009 at 11:01 AM, Corrado <ct529 at york.ac.uk> wrote:
> Dear Octave friends,
>
> here is the benchmark comparison between Octave 322 on Kubuntu 904 64 bit
> compiled from source with and without the --march=native flag set.
>
> It seem that with the new gcc the flag -march-native yields higher
> performances.
>
> What do you think?
>

The most relevant is the for loop benchmark. In the others, the bulk
of the work is done in libraries and -march=native brings you little
benefits. Did you include -march=native in FFLAGS?

> PS: I believe recompiling lapack and blas with -march=native may lead to even
> higher performances, but I have found lapack / blas recompilation rather
> troublesome.
>

It's not that hard, but regarding the BLAS, it makes little sense to
do so with the reference BLAS implementation. A tuned implementation
such as ATLAS or some of the non-free ones will bring you much better
speed.



-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From ct529 at york.ac.uk  Mon Sep 21 04:29:37 2009
From: ct529 at york.ac.uk (Corrado)
Date: Mon, 21 Sep 2009 10:29:37 +0100
Subject: Compiling with and without -march=native: benchmarks
In-Reply-To: <69d8d540909210221k6146b027rf404b22dfca1d468@mail.gmail.com>
References: <200909211001.50518.ct529@york.ac.uk>
	<69d8d540909210221k6146b027rf404b22dfca1d468@mail.gmail.com>
Message-ID: <200909211029.37403.ct529@york.ac.uk>

On Monday 21 September 2009 10:21:46 Jaroslav Hajek wrote:
> benefits. Did you include -march=native in FFLAGS?

yes

> It's not that hard, but regarding the BLAS, it makes little sense to
> do so with the reference BLAS implementation. A tuned implementation
> such as ATLAS or some of the non-free ones will bring you much better
> speed.

yes, I have atlas. I try not to use non free (as in speech) software. So these 
are benchmarks carried out with purely free software / open source software.

Best,
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk


From highegg at gmail.com  Mon Sep 21 04:44:18 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Mon, 21 Sep 2009 11:44:18 +0200
Subject: Compiling with and without -march=native: benchmarks
In-Reply-To: <200909211029.37403.ct529@york.ac.uk>
References: <200909211001.50518.ct529@york.ac.uk>
	<69d8d540909210221k6146b027rf404b22dfca1d468@mail.gmail.com>
	<200909211029.37403.ct529@york.ac.uk>
Message-ID: <69d8d540909210244q13c26a86m3ce0663129e0ad46@mail.gmail.com>

On Mon, Sep 21, 2009 at 11:29 AM, Corrado <ct529 at york.ac.uk> wrote:
> On Monday 21 September 2009 10:21:46 Jaroslav Hajek wrote:
>> benefits. Did you include -march=native in FFLAGS?
>
> yes
>
>> It's not that hard, but regarding the BLAS, it makes little sense to
>> do so with the reference BLAS implementation. A tuned implementation
>> such as ATLAS or some of the non-free ones will bring you much better
>> speed.
>
> yes, I have atlas. I try not to use non free (as in speech) software. So these
> are benchmarks carried out with purely free software / open source software.
>

A self-compiled ATLAS? That is typically faster, because it will adapt
to your machine better.
In general, ATLAS will be able to discover your machine's capabilities
itself, so there's little need to specify performance flags.


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

From przemek at nist.gov  Mon Sep 21 07:21:37 2009
From: przemek at nist.gov (Przemek Klosowski)
Date: Mon, 21 Sep 2009 08:21:37 -0400
Subject: Some Octave newbie questions regarding number format and matrices
In-Reply-To: <9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>
References: <25522543.post@talk.nabble.com>
	<9543b3a40909192306v237f6f8bu7768e6a3930cc9f@mail.gmail.com>
Message-ID: <4AB76FD1.7050306@nist.gov>

On 09/20/2009 02:06 AM, Jordi Guti?rrez Hermos wrote:
> 2009/9/19 hdfrango<hdfrango at gmail.com>:
>> 1. is there any way to force octave to express numbers in fraction form, to
>> avoid loss of precision?
>
> If you don't want to lose precision, you need a symbolic CAS, like
> Sage or Maxima, not Octave.

True in general, but you can use the command 'format rat' to at least
output the numbers as fractions:

octave:7> 1/3
ans = 1/3

octave:8> pi
ans = 355/113

There's also the rat() function that calculates rational approximations
to numbers to a given tolerance.


From WGHM.Janssen at science.ru.nl  Mon Sep 21 03:56:38 2009
From: WGHM.Janssen at science.ru.nl (Wim Janssen)
Date: Mon, 21 Sep 2009 10:56:38 +0200
Subject: Octave and Scientific Linux
Message-ID: <4AB73FC6.9010905@science.ru.nl>

Hello Folks,

As a system-administrator I got several questions and requests
to have the Octave-package installed on our workstation-computers.
I think that's a very good suggestion and  I-am very positive about this.
But unfortunately I was not able to install this package on our 
workstations.
As you know Scientific Linux is very close to the RedHat Distribution,
but it is made more mature to face the high demands on Cern where this
Linux-distribution originates.

Here is my try-to-have-it installed session:
  - Installation from the source-rpm
    [root at schiedam octave-modules]# rpmbuild --rebuild --target i686 
--define \
    > "kernel_topdir /lib/modules/2.6.18-128.1.6.el5/build" \
    > /root/octave-modules/octave-3.0.1-2.el5.src.rpm
    Installing /root/octave-modules/octave-3.0.1-2.el5.src.rpm
    warning: InstallSourcePackage: Header V3 DSA signature: NOKEY, key 
ID 217521    warning: user mockbuild does not exist - using root
    warning: group mockbuild does not exist - using root
    warning: user mockbuild does not exist - using root
    warning: group mockbuild does not exist - using root
    Building target platforms: i686
    Building for target i686
    error: Failed build dependencies:
         hdf5-devel is needed by octave-3.0.1-2.i686
   [root at schiedam octave-modules]#

- I don't know where to find the proper hdf5-devel module for my 
SL-distribution.

- Building Octave completely from source.
  The ./configure runs okay.
   The 'make' runs very long without errors but than stumbles and stops 
a few inches
   before reaching the finish-line. :-(

Help is very Welcome

Kind regards,
                                                        Wim Janssen

-- 
 Radboud Universiteit Nijmegen  Phone: +31(0)24-3652097  [ Wim Janssen ]
 Hoge Energie Fysica HG03.822   Fax: +31(0)24-3652191     Systeembeheer
 Toernooiveld 1            E-mail: WGHM.Janssen at science.ru.nl
 6525 ED  Nijmegen            URL: http://www.hef.ru.nl


From Bart.Vandewoestyne at telenet.be  Mon Sep 21 12:20:00 2009
From: Bart.Vandewoestyne at telenet.be (Bart Vandewoestyne)
Date: Mon, 21 Sep 2009 19:20:00 +0200
Subject: startup with .octaverc on Windows Vista
Message-ID: <20090921172000.GA8578@simba>

Hello,

One of our sysadmins installed Octave 3.2.2 on Windows Vista so
that we can use it as a Matlab replacement for our student's
laboratory sessions.

I went to test the setup today, and i figured out that the
.octaverc file did not get read from the student's H:-drive,
which is the network drive containing their home directory.

I noticed that the HOME environment variable is not set, but the
HOMEDRIVE points to C: and HOMEPATH to \Users\myusername

So if i place the .octaverc file in C:\Users\myusername then it
gets read.  So i can ask students each time they login, to copy
the .octaverc file to C:\Users\myusername but that seems
annoying.  It would be better if they only created .octaverc
once in their H:-drive homedirectory.  So therefore, we will have
to make sure that Octave searches for the .octaverc file in the
H:\ path.  While googling, i found

http://www.nabble.com/Home-directory-in-octave-on-windows--td23772097.html

and it seems like there exist several environment variables that
are candidates, namely:

1) HOME

2) HOMEDRIVE + HOMEPATH

3) USERPROFILE

My questions are:

1) What environment variables does Octave read and use to search
for an .octaverc file?

2) Considering that I am looking for a clean and long-term solution,
   what would be the best environment variable to use on Windows
   Vista?

Thanks for any advice,
Bart

-- 
	"Share what you know.  Learn what you don't."

From ar63 at evansville.edu  Mon Sep 21 13:43:26 2009
From: ar63 at evansville.edu (Richardson, Anthony)
Date: Mon, 21 Sep 2009 13:43:26 -0500
Subject: startup with .octaverc on Windows Vista
In-Reply-To: <20090921172000.GA8578@simba>
References: <20090921172000.GA8578@simba>
Message-ID: <2318D9796D5B9D41B85612B5C87F224002350E5A@UEEXCHANGE.evansville.edu>

We have a similar setup.   We set HOME in the
C:/octave/3.22.2_gcc-4.3.0/share/octave/site/m/startup/octaverc
file, with the following lines:

setenv('HOME','h:\');
cd ~;

This allows the octave HOME env variable to be set differently 
than a Vista global HOME environment variable.  An .octaverc 
file in h:\ is still executed.  Setting HOME from Windows
should work too, but we have other software that uses the HOME
environment variable and requires that it be set to something
else.

Tony Richardson

> -----Original Message-----
> From: Bart Vandewoestyne [mailto:Bart.Vandewoestyne at telenet.be]
> Sent: Monday, September 21, 2009 12:20 PM
> To: help-octave at octave.org
> Subject: startup with .octaverc on Windows Vista
> 
> Hello,
> 
> One of our sysadmins installed Octave 3.2.2 on Windows Vista so
> that we can use it as a Matlab replacement for our student's
> laboratory sessions.
> 
> I went to test the setup today, and i figured out that the
> .octaverc file did not get read from the student's H:-drive,
> which is the network drive containing their home directory.
> 
> I noticed that the HOME environment variable is not set, but the
> HOMEDRIVE points to C: and HOMEPATH to \Users\myusername
> 
> So if i place the .octaverc file in C:\Users\myusername then it
> gets read.  So i can ask students each time they login, to copy
> the .octaverc file to C:\Users\myusername but that seems
> annoying.  It would be better if they only created .octaverc
> once in their H:-drive homedirectory.  So therefore, we will have
> to make sure that Octave searches for the .octaverc file in the
> H:\ path.  While googling, i found
> 
> http://www.nabble.com/Home-directory-in-octave-on-windows--
> td23772097.html
> 
> and it seems like there exist several environment variables that
> are candidates, namely:
> 
> 1) HOME
> 
> 2) HOMEDRIVE + HOMEPATH
> 
> 3) USERPROFILE
> 
> My questions are:
> 
> 1) What environment variables does Octave read and use to search
> for an .octaverc file?
> 
> 2) Considering that I am looking for a clean and long-term solution,
>    what would be the best environment variable to use on Windows
>    Vista?
> 
> Thanks for any advice,
> Bart
> 
> --
> 	"Share what you know.  Learn what you don't."
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave


From ottar-s at hotmail.com  Tue Sep 22 11:29:28 2009
From: ottar-s at hotmail.com (=?Windows-1252?B?T3R0YXIgU+Z0cmU=?=)
Date: Tue, 22 Sep 2009 18:29:28 +0200
Subject: Plotting problems
Message-ID: <SNT113-W54962D26CBF3E9FCF06CA8EDDC0@phx.gbl>


Hi,
So far I have worked with matrices, but now I want to make plots. That?s not easy.
When I try the command:
plot(x, y);
I get a series of answers. In the first lines I read: ?sh: gnuplot: not found?.
When there are a lot of errors, mostly with if-statements, but also a for-loop.
Errors are one time located to:
?/octave/3.0.0/m/miscellanous/compare_versions.m
And three times located to
?/octave/3.0.0/m/plot/drawnow.m
 
I have heard that I need to download a plot packet called V3.0. I also tried to upgrade octave.
When I try to upgrade octave with the command
?sudo apt-get build-dep octave?
I get the message:
 ?Can?t find a source code packet for octave?
 
Thanks for any advice,
Ottar S?tre 		 	   		  
_________________________________________________________________
Drag n? drop?Get easy photo sharing with Windows Live? Photos.

http://www.microsoft.com/windows/windowslive/products/photos.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090922/71768bc3/attachment.html 

From soren at hauberg.org  Tue Sep 22 11:41:48 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Tue, 22 Sep 2009 18:41:48 +0200
Subject: Plotting problems
In-Reply-To: <SNT113-W54962D26CBF3E9FCF06CA8EDDC0@phx.gbl>
References: <SNT113-W54962D26CBF3E9FCF06CA8EDDC0@phx.gbl>
Message-ID: <1253637708.4423.2.camel@sh-t400>

tir, 22 09 2009 kl. 18:29 +0200, skrev Ottar S?tre:
> I get a series of answers. In the first lines I read: ?sh: gnuplot:
> not found?.

Sounds like you need to install 'gnuplot'. If it is already installed,
perhaps you should look at the 'gnuplot_binary' function in Octave.

S?ren


From gbarrick at walsh.edu  Tue Sep 22 11:44:59 2009
From: gbarrick at walsh.edu (George Barrick)
Date: Tue, 22 Sep 2009 12:44:59 -0400
Subject: Feature request.
Message-ID: <4B30AD0BED19E842AAE88DC39736493123CC00BD99@mail3.walsh.edu>

Subject: feature request.

                            2009.09.22.12:15:56 EDT

Hello GNU Octave folks,

      I'd like to request that the 'position' property
of figure() windows be implemented in Octave.

      I typically like to have fine control over the
diagrams that I exhibit for my students.  I _can_ get
control over the proportions of the graphing window
with print statements like:

print -djpeg "-S384,384" f44posn.jpg ;

in my octave scripts.  However, it would be nice if
the X11-display of the graph precisely mimicked the
print output whenever I demand of the octave-interpreter
a window having particular proportions.

     With my current set-up:

     WinXP Pro. SP3 ver. 5.1 build 2600
     CygWin 1.7.0-61
     Xorg-x11 1.6.3-1
     Octave 3.2.3-1
     Gnuplot 4.2.4-1

the octave-interpreter ignores me when I set the
'postion' property of my figure().  It always creates
an x11-window that is 640x480.

     If this is already done in the octave-3.3 development,
forgive my ignorance.  I haven't had time to keep up
with the development code in the mercurial repository.

     Anyone who is interested in such a feature can
see whether this behavior is reproduced on their local
set-up.  At the end of this post is my script for
a parabola where I ask octave to create a square window.
Note that the statement in my script:

set( kdle,'NextPlot','add')                     ;

is important.  If I run the script without that line,
octave seems to re-spawn the axes upon further graphics
calls, and does whatever it pleases with the figure and
axis properties.

Thanks,

George                         gbarrick at walsh.edu
                               http://www.gbarrick.net


% <file: fall44.m>
%********D%********D%********D%********D%********D%********D%********D%*
%%  Set up graphing parameters.  Set vectors that state
%%    window position and size.

winXst     =  256 ;
winYst     =  256 ;
winWdt     =  256 ;
winHgt     =  256 ;

winPos  = [ winXst winYst winWdt winHgt ] ;

%********D%********D%********D%********D%********D%********D%********D%*
%%  Set the vector that gives area inside the window where
%%    axes will be drawn. All locations are stated relative
%%    to the window position.  Use window width and height
%%    as denominators.

win    = [ winWdt winHgt winWdt winHgt ] ;

plotWdt     =   210	;
plotHgt     =   210	;
plotXst     =    38	;
plotYst     =    38	;

plotPos     =  [ plotXst plotYst plotWdt plotHgt ] ./ win ;

%%  Set limits for the axes.

xlims    =	[  0.00, 2.10 ]	;
ylims    =	[  0.00, 2.10 ]	;

%********D%********D%********D%********D%********D%********D%********D%***f**
%%  The graphics functions figure() and axes() are called using
%%    the settings previously written up.  Use "hdle" as the graphics
%%    handle for the figure window.

hdle = figure()			;
set( hdle,'Position',winPos )	;

%% Spawn axes first, then re-configure them.
kdle = axes() ;
set( kdle,'Position',plotPos )			;
set( kdle,'XLim',xlims,'YLim',ylims )		;
set( kdle,'NextPlot','add')                     ;

%%  Create the axis labels.

kt3 = xlabel( 'x(t)' )                          ;
ku3 = ylabel( 'y(t)' )                          ;

%%  Create the graph.

xrow  = [  0.00 : 0.02 : 0.62  ] ;
xvals = [  xrow , 2*sqrt(5)/7  ] ;

wunz  = ones( size( xvals ) )    ;

gee   =  9.80                    ;

yvals = ( -0.5 * gee * ( xvals.^2 )) + ( 2.0 * wunz ) ;

l1    = plot( xvals, yvals, 'k-' );
set(l1,'LineWidth',2) ;


hndle = [ hdle; kdle; kt3; ku3; l1 ];

%% print -djpeg "-S384,384" f44posn.jpg;
%</file>



From shermanj at umd.edu  Tue Sep 22 16:44:49 2009
From: shermanj at umd.edu (James Sherman Jr.)
Date: Tue, 22 Sep 2009 17:44:49 -0400
Subject: Feature request.
In-Reply-To: <4B30AD0BED19E842AAE88DC39736493123CC00BD99@mail3.walsh.edu>
References: <4B30AD0BED19E842AAE88DC39736493123CC00BD99@mail3.walsh.edu>
Message-ID: <6800c3890909221444i73d5114dtbbd4ba9bb7bb004@mail.gmail.com>

I believe that this is/was a problem with gnuplot rather than with octave,
but I believe that this has been fixed in the most recent version.

http://www.nabble.com/gnuplot--%3E-octave-communication-of-figure-position-to21736426.html#a21736426
http://www.nabble.com/-changeset--figure-position-property-support-for-gnuplot---x11-to21260890.html#a21260890
http://www.nabble.com/gnuplot%2C-figure.position%2C---the-moving-window-to22069067.html#a22069067
http://www.nabble.com/Figure-Position-to24790735.html#a24790735

I can't test it currently, but this is what I found from a quick search of
the archives.

On Tue, Sep 22, 2009 at 12:44 PM, George Barrick <gbarrick at walsh.edu> wrote:

> Subject: feature request.
>
>                            2009.09.22.12:15:56 EDT
>
> Hello GNU Octave folks,
>
>      I'd like to request that the 'position' property
> of figure() windows be implemented in Octave.
>
>      I typically like to have fine control over the
> diagrams that I exhibit for my students.  I _can_ get
> control over the proportions of the graphing window
> with print statements like:
>
> print -djpeg "-S384,384" f44posn.jpg ;
>
> in my octave scripts.  However, it would be nice if
> the X11-display of the graph precisely mimicked the
> print output whenever I demand of the octave-interpreter
> a window having particular proportions.
>
>     With my current set-up:
>
>     WinXP Pro. SP3 ver. 5.1 build 2600
>     CygWin 1.7.0-61
>     Xorg-x11 1.6.3-1
>     Octave 3.2.3-1
>     Gnuplot 4.2.4-1
>
> the octave-interpreter ignores me when I set the
> 'postion' property of my figure().  It always creates
> an x11-window that is 640x480.
>
>     If this is already done in the octave-3.3 development,
> forgive my ignorance.  I haven't had time to keep up
> with the development code in the mercurial repository.
>
>     Anyone who is interested in such a feature can
> see whether this behavior is reproduced on their local
> set-up.  At the end of this post is my script for
> a parabola where I ask octave to create a square window.
> Note that the statement in my script:
>
> set( kdle,'NextPlot','add')                     ;
>
> is important.  If I run the script without that line,
> octave seems to re-spawn the axes upon further graphics
> calls, and does whatever it pleases with the figure and
> axis properties.
>
> Thanks,
>
> George                         gbarrick at walsh.edu
>                               http://www.gbarrick.net
>
>
> % <file: fall44.m>
> %********D%********D%********D%********D%********D%********D%********D%*
> %%  Set up graphing parameters.  Set vectors that state
> %%    window position and size.
>
> winXst     =  256 ;
> winYst     =  256 ;
> winWdt     =  256 ;
> winHgt     =  256 ;
>
> winPos  = [ winXst winYst winWdt winHgt ] ;
>
> %********D%********D%********D%********D%********D%********D%********D%*
> %%  Set the vector that gives area inside the window where
> %%    axes will be drawn. All locations are stated relative
> %%    to the window position.  Use window width and height
> %%    as denominators.
>
> win    = [ winWdt winHgt winWdt winHgt ] ;
>
> plotWdt     =   210     ;
> plotHgt     =   210     ;
> plotXst     =    38     ;
> plotYst     =    38     ;
>
> plotPos     =  [ plotXst plotYst plotWdt plotHgt ] ./ win ;
>
> %%  Set limits for the axes.
>
> xlims    =      [  0.00, 2.10 ] ;
> ylims    =      [  0.00, 2.10 ] ;
>
>
> %********D%********D%********D%********D%********D%********D%********D%***f**
> %%  The graphics functions figure() and axes() are called using
> %%    the settings previously written up.  Use "hdle" as the graphics
> %%    handle for the figure window.
>
> hdle = figure()                 ;
> set( hdle,'Position',winPos )   ;
>
> %% Spawn axes first, then re-configure them.
> kdle = axes() ;
> set( kdle,'Position',plotPos )                  ;
> set( kdle,'XLim',xlims,'YLim',ylims )           ;
> set( kdle,'NextPlot','add')                     ;
>
> %%  Create the axis labels.
>
> kt3 = xlabel( 'x(t)' )                          ;
> ku3 = ylabel( 'y(t)' )                          ;
>
> %%  Create the graph.
>
> xrow  = [  0.00 : 0.02 : 0.62  ] ;
> xvals = [  xrow , 2*sqrt(5)/7  ] ;
>
> wunz  = ones( size( xvals ) )    ;
>
> gee   =  9.80                    ;
>
> yvals = ( -0.5 * gee * ( xvals.^2 )) + ( 2.0 * wunz ) ;
>
> l1    = plot( xvals, yvals, 'k-' );
> set(l1,'LineWidth',2) ;
>
>
> hndle = [ hdle; kdle; kt3; ku3; l1 ];
>
> %% print -djpeg "-S384,384" f44posn.jpg;
> %</file>
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090922/b889abac/attachment.html 

From Bart.Vandewoestyne at telenet.be  Thu Sep 24 01:19:26 2009
From: Bart.Vandewoestyne at telenet.be (Bart Vandewoestyne)
Date: Thu, 24 Sep 2009 08:19:26 +0200
Subject: startup with .octaverc on Windows Vista
In-Reply-To: <2318D9796D5B9D41B85612B5C87F224002350E5A@UEEXCHANGE.evansville.edu>
References: <20090921172000.GA8578@simba>
	<2318D9796D5B9D41B85612B5C87F224002350E5A@UEEXCHANGE.evansville.edu>
Message-ID: <20090924061926.GA16446@simba>

On Mon, Sep 21, 2009 at 01:43:26PM -0500, Richardson, Anthony wrote:
> We have a similar setup.   We set HOME in the
> C:/octave/3.22.2_gcc-4.3.0/share/octave/site/m/startup/octaverc
> file, with the following lines:
> 
> setenv('HOME','h:\');
> cd ~;
> 
> This allows the octave HOME env variable to be set differently 
> than a Vista global HOME environment variable.  An .octaverc 
> file in h:\ is still executed.  Setting HOME from Windows
> should work too, but we have other software that uses the HOME
> environment variable and requires that it be set to something
> else.
> 
> Tony Richardson

Thanks for your reply, Tony.  In our computer-rooms, we are doing
it now the way you describe, and everything works nicely.  The
Octave desktop-icon is now shining on all our student-computers
:-)

I do have another question related to the installer though...

For our sysadmins, it would be comfortable if they can do a
silent (unattended) install.  For as far as I can see, Octave
uses the NSIS installer system (http://nsis.sourceforge.net) and
it has an option /S with which one can do a silent install (see
for example http://unattended.sourceforge.net/installers.php)

However, if one performs a silent install, then apparently the
installation is done with the *default* options, that is, no
Octave-Forge packages are installed and one cannot select another
Atlas library.

Is it somehow possible to do a silent (unattended) install *with*
specifying that Octave-Forge should also be installed?  Does the
NSIS installer system allow a user to specify certain
installation options somehow?

Thanks,
Bart

-- 
	"Share what you know.  Learn what you don't."

From dryeyedman at gmail.com  Thu Sep 24 12:55:05 2009
From: dryeyedman at gmail.com (Enrique Sahagun)
Date: Thu, 24 Sep 2009 19:55:05 +0200
Subject: Octave GUI tools for Windows XP
Message-ID: <4ABBB279.4050108@gmail.com>

Hi all:

I'm new in Octave and I have learnt how to make some interesting 
calculations with it. Now I'm working in an image tracking software (to 
follow an object on a video sequence). I would like this program to have 
an interface and I know that with MatLab it is possible to make friendly 
interfaces. And I was wondering, IT IS POSSIBLE TO CREATE AN INTERFACE 
FOR YOUR SCRIPT IN OCTAVE?.

PD: I'm using octave over windows XP.

Thanks in advance.

From primozz.peterlin at gmail.com  Fri Sep 25 02:38:43 2009
From: primozz.peterlin at gmail.com (Primoz PETERLIN)
Date: Fri, 25 Sep 2009 09:38:43 +0200
Subject: Regional maxima of an image
Message-ID: <d8daf73b0909250038g1d4ad172macd223d3f9ccabf3@mail.gmail.com>

Hello everybody,

Does Octave already has a function which returns a map of regional
maxima (I am looking for something like imregionalmax() from Matlab
Image Processing Toolbox), or would it be worthwhile to write one?

All the best,
Primoz

-- 
Primo? Peterlin,   In?titut za biofiziko, Med. fakulteta, Univerza v Ljubljani
Lipi?eva 2, SI-1000 Ljubljana, Slovenija.        primoz.peterlin at mf.uni-lj.si
Tel +386-1-5437612, fax +386-1-4315127,  http://biofiz.mf.uni-lj.si/~peterlin/
F8021D69 OpenPGP fingerprint: CB 6F F1 EE D9 67 E0 2F  0B 59 AF 0D 79 56 19 0F


From soren at hauberg.org  Fri Sep 25 02:45:52 2009
From: soren at hauberg.org (=?ISO-8859-1?Q?S=F8ren?= Hauberg)
Date: Fri, 25 Sep 2009 09:45:52 +0200
Subject: Regional maxima of an image
In-Reply-To: <d8daf73b0909250038g1d4ad172macd223d3f9ccabf3@mail.gmail.com>
References: <d8daf73b0909250038g1d4ad172macd223d3f9ccabf3@mail.gmail.com>
Message-ID: <1253864752.4625.1.camel@sh-t400>

fre, 25 09 2009 kl. 09:38 +0200, skrev Primoz PETERLIN:
> Does Octave already has a function which returns a map of regional
> maxima (I am looking for something like imregionalmax() from Matlab
> Image Processing Toolbox), or would it be worthwhile to write one?

Does the 'immaximas' from the 'image' package do what you want? If not,
could you then describe what 'imregionalmax' does? In either case, I
guess it would be nice to have 'imregionalmax' for compatibility.

S?ren


From drvine at gmail.com  Fri Sep 25 03:37:38 2009
From: drvine at gmail.com (Troy Vine)
Date: Fri, 25 Sep 2009 10:37:38 +0200
Subject: Unable to get transparent region using patch(...)
Message-ID: <1E9E8320-8AEF-4A40-A217-55F373609188@gmail.com>

Hi,

I am using octave 3.2.2 on OS X 10.5.8 installed using the standard  
installer.  Using the code below, I attempted to create a shaded  
region transparent to the plotted line.  With both X11 and AquaTerm  
(version 1.0.1) the shaded region is opaque to the plot underneath.   
If I execute the command

 > get(p, 'facealpha')

afterwards then I get 0.2 as expected.  What am I doing wrong?

Many thanks,

Troy



 > x = [0:pi/12:2*pi];
 > plot(x, sin(x))
 > xp = [2 2 3 3];
 > yp = [ylim fliplr(ylim)];
 > p = patch(xp,yp, 'r');
 > set(p,'facealpha',0.2);


From jon_goitia at hotmail.com  Thu Sep 24 15:24:54 2009
From: jon_goitia at hotmail.com (Jon Goitia)
Date: Thu, 24 Sep 2009 20:24:54 +0000
Subject: Installing problems
Message-ID: <COL107-W49B417DC3D8BA28AD7791FE6DA0@phx.gbl>


Hi,

I wanted to install octave, but I'm not sure on how to do it. I opened the FTP page and downloaded the octave-3.2.2.tar.gz compressed file. I extracted all the files, but I don't know how to install the program. I am using Windows XP. What do I need to do?

 

Thanks
 		 	   		  
_________________________________________________________________
Nuevo Windows Live, un mundo lleno de posibilidades. Desc?brelo.
http://www.microsoft.com/windows/windowslive/default.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090924/846888e1/attachment.html 

From dryeyedman at gmail.com  Fri Sep 25 06:15:24 2009
From: dryeyedman at gmail.com (Enrique Sahagun)
Date: Fri, 25 Sep 2009 13:15:24 +0200
Subject: Octave GUI tools for Windows XP
In-Reply-To: <4ABCA3B7.4050704@acc.umu.se>
References: <4ABBB279.4050108@gmail.com> <4ABCA3B7.4050704@acc.umu.se>
Message-ID: <4ABCA64C.1060409@gmail.com>

David Grundberg escribi?:
> Enrique Sahagun wrote:
>> Hi all:
>>
>> I'm new in Octave and I have learnt how to make some interesting 
>> calculations with it. Now I'm working in an image tracking software 
>> (to follow an object on a video sequence). I would like this program 
>> to have an interface and I know that with MatLab it is possible to 
>> make friendly interfaces. And I was wondering, IT IS POSSIBLE TO 
>> CREATE AN INTERFACE FOR YOUR SCRIPT IN OCTAVE?.
>>
>> PD: I'm using octave over windows XP.
>>
>> Thanks in advance.
>> _______________________________________________
>> Help-octave mailing list
>> Help-octave at octave.org
>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>>   
> I made a GUI app using Python and GTK+. In that application, I used 
> Pytave (Octave module for Python) to execute Octave code. I'm the 
> Pytave maintainer, and I haven't heard anyone building Pytave for 
> Windows though. Any help with regards to that is welcome.
>
> I suggest you don't use all caps (LIKE THIS), in my opinion it looks 
> obnoxious.
>
> hth,
> David
>
Yeah, my approach at this moment is to learn some Python to try to 
develope an app to call octave from it. At the moment I'm just trying to 
taste how difficult would be.

And sorry for the caps, I was only trying to underlying the question but 
you are right.

Thanks
Enrique

From jordigh at gmail.com  Fri Sep 25 06:19:00 2009
From: jordigh at gmail.com (=?UTF-8?Q?Jordi_Guti=C3=A9rrez_Hermoso?=)
Date: Fri, 25 Sep 2009 06:19:00 -0500
Subject: Installing problems
In-Reply-To: <COL107-W49B417DC3D8BA28AD7791FE6DA0@phx.gbl>
References: <COL107-W49B417DC3D8BA28AD7791FE6DA0@phx.gbl>
Message-ID: <9543b3a40909250419r64fe30b2wae52173db00f9434@mail.gmail.com>

2009/9/24 Jon Goitia <jon_goitia at hotmail.com>:
> I wanted to install octave,
[...]
> I am using  Windows XP. What do I need to do?

You downloaded the source code. Unless you want to compile it
yourself, you probably want to download the windows installer from
here instead:

     http://octave.sourceforge.net/

Hope this helps,
- Jordi G. H.

From individ at acc.umu.se  Fri Sep 25 06:04:23 2009
From: individ at acc.umu.se (David Grundberg)
Date: Fri, 25 Sep 2009 13:04:23 +0200
Subject: Octave GUI tools for Windows XP
In-Reply-To: <4ABBB279.4050108@gmail.com>
References: <4ABBB279.4050108@gmail.com>
Message-ID: <4ABCA3B7.4050704@acc.umu.se>

Enrique Sahagun wrote:
> Hi all:
>
> I'm new in Octave and I have learnt how to make some interesting 
> calculations with it. Now I'm working in an image tracking software (to 
> follow an object on a video sequence). I would like this program to have 
> an interface and I know that with MatLab it is possible to make friendly 
> interfaces. And I was wondering, IT IS POSSIBLE TO CREATE AN INTERFACE 
> FOR YOUR SCRIPT IN OCTAVE?.
>
> PD: I'm using octave over windows XP.
>
> Thanks in advance.
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>   
I made a GUI app using Python and GTK+. In that application, I used 
Pytave (Octave module for Python) to execute Octave code. I'm the Pytave 
maintainer, and I haven't heard anyone building Pytave for Windows 
though. Any help with regards to that is welcome.

I suggest you don't use all caps (LIKE THIS), in my opinion it looks 
obnoxious.

hth,
David

From highegg at gmail.com  Fri Sep 25 06:34:56 2009
From: highegg at gmail.com (Jaroslav Hajek)
Date: Fri, 25 Sep 2009 13:34:56 +0200
Subject: Octave GUI tools for Windows XP
In-Reply-To: <4ABCA64C.1060409@gmail.com>
References: <4ABBB279.4050108@gmail.com> <4ABCA3B7.4050704@acc.umu.se>
	<4ABCA64C.1060409@gmail.com>
Message-ID: <69d8d540909250434m423dc3ectaeb5db6505ec5cc4@mail.gmail.com>

On Fri, Sep 25, 2009 at 1:15 PM, Enrique Sahagun <dryeyedman at gmail.com> wrote:
> David Grundberg escribi?:
>> Enrique Sahagun wrote:
>>> Hi all:
>>>
>>> I'm new in Octave and I have learnt how to make some interesting
>>> calculations with it. Now I'm working in an image tracking software
>>> (to follow an object on a video sequence). I would like this program
>>> to have an interface and I know that with MatLab it is possible to
>>> make friendly interfaces. And I was wondering, IT IS POSSIBLE TO
>>> CREATE AN INTERFACE FOR YOUR SCRIPT IN OCTAVE?.
>>>
>>> PD: I'm using octave over windows XP.
>>>
>>> Thanks in advance.
>>> _______________________________________________
>>> Help-octave mailing list
>>> Help-octave at octave.org
>>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>>>
>> I made a GUI app using Python and GTK+. In that application, I used
>> Pytave (Octave module for Python) to execute Octave code. I'm the
>> Pytave maintainer, and I haven't heard anyone building Pytave for
>> Windows though. Any help with regards to that is welcome.
>>
>> I suggest you don't use all caps (LIKE THIS), in my opinion it looks
>> obnoxious.
>>
>> hth,
>> David
>>
> Yeah, my approach at this moment is to learn some Python to try to
> develope an app to call octave from it. At the moment I'm just trying to
> taste how difficult would be.
>

With Pytave, it should be easier than, for instance, communicating
through the subprocess module. More efficient as well.
Pretty good for embedding Octave computations in a larger Python
framework (for instance, a web app).

> And sorry for the caps, I was only trying to underlying the question but
> you are right.
>
> Thanks
> Enrique
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>


-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz


From dryeyedman at gmail.com  Fri Sep 25 06:39:47 2009
From: dryeyedman at gmail.com (Enrique Sahagun)
Date: Fri, 25 Sep 2009 13:39:47 +0200
Subject: Octave GUI tools for Windows XP
In-Reply-To: <69d8d540909250434m423dc3ectaeb5db6505ec5cc4@mail.gmail.com>
References: <4ABBB279.4050108@gmail.com> <4ABCA3B7.4050704@acc.umu.se>	
	<4ABCA64C.1060409@gmail.com>
	<69d8d540909250434m423dc3ectaeb5db6505ec5cc4@mail.gmail.com>
Message-ID: <4ABCAC03.2010902@gmail.com>

Jaroslav Hajek escribi?:
> On Fri, Sep 25, 2009 at 1:15 PM, Enrique Sahagun <dryeyedman at gmail.com> wrote:
>   
>> David Grundberg escribi?:
>>     
>>> Enrique Sahagun wrote:
>>>       
>>>> Hi all:
>>>>
>>>> I'm new in Octave and I have learnt how to make some interesting
>>>> calculations with it. Now I'm working in an image tracking software
>>>> (to follow an object on a video sequence). I would like this program
>>>> to have an interface and I know that with MatLab it is possible to
>>>> make friendly interfaces. And I was wondering, IT IS POSSIBLE TO
>>>> CREATE AN INTERFACE FOR YOUR SCRIPT IN OCTAVE?.
>>>>
>>>> PD: I'm using octave over windows XP.
>>>>
>>>> Thanks in advance.
>>>> _______________________________________________
>>>> Help-octave mailing list
>>>> Help-octave at octave.org
>>>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>>>>
>>>>         
>>> I made a GUI app using Python and GTK+. In that application, I used
>>> Pytave (Octave module for Python) to execute Octave code. I'm the
>>> Pytave maintainer, and I haven't heard anyone building Pytave for
>>> Windows though. Any help with regards to that is welcome.
>>>
>>> I suggest you don't use all caps (LIKE THIS), in my opinion it looks
>>> obnoxious.
>>>
>>> hth,
>>> David
>>>
>>>       
>> Yeah, my approach at this moment is to learn some Python to try to
>> develope an app to call octave from it. At the moment I'm just trying to
>> taste how difficult would be.
>>
>>     
>
> With Pytave, it should be easier than, for instance, communicating
> through the subprocess module. More efficient as well.
> Pretty good for embedding Octave computations in a larger Python
> framework (for instance, a web app).
>
>   
>> And sorry for the caps, I was only trying to underlying the question but
>> you are right.
>>
>> Thanks
>> Enrique
>> _______________________________________________
>> Help-octave mailing list
>> Help-octave at octave.org
>> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>>
>>     
>
>
>   
Ok, but does Pytave work in Windows XP? I work on both Unix systems and 
Windows but for this particular project I need  to create a friendly 
interface on Windows, able to run (and write) Octave.

From mbmiller+l at gmail.com  Fri Sep 25 16:02:46 2009
From: mbmiller+l at gmail.com (Mike Miller)
Date: Fri, 25 Sep 2009 16:02:46 -0500 (CDT)
Subject: noncentral F cdf
Message-ID: <alpine.DEB.2.00.0909251601220.20314@taxa.psych.umn.edu>

Do we have a function that can compute the noncentral F cdf?

I'm using the Octave package on Ubuntu and it doesn't look like I have it 
but it might be out there somewhere else.

Thanks in advance!

Mike

From hellyj at ucsd.edu  Fri Sep 25 19:49:38 2009
From: hellyj at ucsd.edu (Helly John)
Date: Fri, 25 Sep 2009 17:49:38 -0700
Subject: Problem with octave/gnuplot/aquaterm on Snow Leopard
In-Reply-To: <A59396EE-76FB-4027-887F-926E497518C0@ucsd.edu>
References: <A59396EE-76FB-4027-887F-926E497518C0@ucsd.edu>
Message-ID: <E59D6485-57DB-44F3-A837-05F1043A5192@ucsd.edu>

Hi.

I'm new to Octave and gnuplot and Aquaterm but have just installed all
of these on my system.  Here's the configuration:

- Mac OS X 10.6.1
- AquaTerm v1.0.1 (1.0.1)
- Octave-3.2.2
- GNUPLOT 4.2 patchlevel 6

When I try to plot in octave this is what happens:
*****************************************************************
octave-3.2.2:1> Y=[1:10]
Y =

     1    2    3    4    5    6    7    8    9   10

octave-3.2.2:2> plot(Y)
Unknown or ambiguous terminal name 'aqua'

gnuplot> set terminal aqua enhanced title "Figure 1" size 560 420
                       ^
          line 0: unknown or ambiguous terminal type; type just 'set
terminal' for a list

          line 0: No terminal defined


gnuplot> plot "-" binary format='%float64' record=10 using ($1):($2)
axes x1y1 title "" with lines linestyle 1 ;
               ^
          line 0: use 'set term' to set terminal type first

octave-3.2.2:3> system ("echo $GNUTERM")
aqua
ans = 0
octave-3.2.2:4>
*****************************************************************

Do you have any suggestions as to what might be wrong?
Thanks.
-----------------------------------------------
John Helly / San Diego Supercomputer Center & Climate, Atmospheric
Science, and Physical Oceanography/Scripps Institution of
Oceanography / UCSD / +01 760 840 8660 mobile / stonesteps (Skype) /
stonesteps7 (iChat) / http://www.sdsc.edu/~hellyj




-----------------------------------------------
John Helly / San Diego Supercomputer Center & Climate, Atmospheric  
Science, and Physical Oceanography/Scripps Institution of  
Oceanography / UCSD / +01 760 840 8660 mobile / stonesteps (Skype) /  
stonesteps7 (iChat) / http://www.sdsc.edu/~hellyj




From tmacchant at yahoo.co.jp  Fri Sep 25 22:41:08 2009
From: tmacchant at yahoo.co.jp (Tatsuro MATSUOKA)
Date: Sat, 26 Sep 2009 12:41:08 +0900 (JST)
Subject: Problem with octave/gnuplot/aquaterm on Snow Leopard
In-Reply-To: <E59D6485-57DB-44F3-A837-05F1043A5192@ucsd.edu>
Message-ID: <20090926034108.50091.qmail@web3813.mail.bbt.yahoo.co.jp>

Hello

My suggestion may not be useful because I'm not a mac user.

Please execute gnuplot itself.
And type 

gnuplot> set terminal aqua enhanced title "Figure 1" size 560 420

from the gnuplot prompt.
 
Does the above works well?

If it does not work, your installing gnuplot was not be successful.
If it works, please wait until mac users' reply.

Regards

Tatsuro
 
--- Helly John> wrote:

> Hi.
> 
> I'm new to Octave and gnuplot and Aquaterm but have just installed all
> of these on my system.  Here's the configuration:
> 
> - Mac OS X 10.6.1
> - AquaTerm v1.0.1 (1.0.1)
> - Octave-3.2.2
> - GNUPLOT 4.2 patchlevel 6
> 
> When I try to plot in octave this is what happens:
> *****************************************************************
> octave-3.2.2:1> Y=[1:10]
> Y =
> 
>      1    2    3    4    5    6    7    8    9   10
> 
> octave-3.2.2:2> plot(Y)
> Unknown or ambiguous terminal name 'aqua'
> 
> gnuplot> set terminal aqua enhanced title "Figure 1" size 560 420
>                        ^
>           line 0: unknown or ambiguous terminal type; type just 'set
> terminal' for a list
> 
>           line 0: No terminal defined
> 
> 
> gnuplot> plot "-" binary format='%float64' record=10 using ($1):($2)
> axes x1y1 title "" with lines linestyle 1 ;
>                ^
>           line 0: use 'set term' to set terminal type first
> 
> octave-3.2.2:3> system ("echo $GNUTERM")
> aqua
> ans = 0
> octave-3.2.2:4>
> *****************************************************************
> 
> Do you have any suggestions as to what might be wrong?
> Thanks.
> -----------------------------------------------
> John Helly / San Diego Supercomputer Center & Climate, Atmospheric
> Science, and Physical Oceanography/Scripps Institution of
> Oceanography / UCSD / +01 760 840 8660 mobile / stonesteps (Skype) /
> stonesteps7 (iChat) / http://www.sdsc.edu/~hellyj
> 
> 
> 
> 
> -----------------------------------------------
> John Helly / San Diego Supercomputer Center & Climate, Atmospheric  
> Science, and Physical Oceanography/Scripps Institution of  
> Oceanography / UCSD / +01 760 840 8660 mobile / stonesteps (Skype) /  
> stonesteps7 (iChat) / http://www.sdsc.edu/~hellyj
> 
> 
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 


--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/

From Thomas.Treichl at gmx.net  Sun Sep 27 05:49:27 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Sun, 27 Sep 2009 12:49:27 +0200
Subject: Problem with octave/gnuplot/aquaterm on Snow Leopard
In-Reply-To: <E59D6485-57DB-44F3-A837-05F1043A5192@ucsd.edu>
References: <A59396EE-76FB-4027-887F-926E497518C0@ucsd.edu>
	<E59D6485-57DB-44F3-A837-05F1043A5192@ucsd.edu>
Message-ID: <4ABF4337.8030405@gmx.net>

Helly John schrieb:
> Hi.
> 
> I'm new to Octave and gnuplot and Aquaterm but have just installed all
> of these on my system.  Here's the configuration:
> 
> - Mac OS X 10.6.1
> - AquaTerm v1.0.1 (1.0.1)
> - Octave-3.2.2
> - GNUPLOT 4.2 patchlevel 6
> 
> When I try to plot in octave this is what happens:
> *****************************************************************
> octave-3.2.2:1> Y=[1:10]
> Y =
> 
>      1    2    3    4    5    6    7    8    9   10
> 
> octave-3.2.2:2> plot(Y)
> Unknown or ambiguous terminal name 'aqua'
> 
> gnuplot> set terminal aqua enhanced title "Figure 1" size 560 420
>                        ^
>           line 0: unknown or ambiguous terminal type; type just 'set
> terminal' for a list
> 
>           line 0: No terminal defined
> 
> 
> gnuplot> plot "-" binary format='%float64' record=10 using ($1):($2)
> axes x1y1 title "" with lines linestyle 1 ;
>                ^
>           line 0: use 'set term' to set terminal type first
> 
> octave-3.2.2:3> system ("echo $GNUTERM")
> aqua
> ans = 0
> octave-3.2.2:4>
> *****************************************************************
> 
> Do you have any suggestions as to what might be wrong?
> Thanks.

Hi,

it looks like that your Gnuplot 4.2.6 has not been compiled against 
libaquaterm.dylib. It's not just that you need to install AquaTerm on your Mac, 
the one who compiles the sources that you have got needs to compile against 
libaquaterm.dylib.
At least I hope that this is your problem. You can check this if you type "set 
terminal" after starting Gnuplot and then check the second line for the "aqua" 
terminal. On my system with Gnuplot 4.2.5 this looks like

   gnuplot> set terminal
   Available terminal types:
              aqua  Interface to graphics terminal server for Mac OS X
               cgm  Computer Graphics Metafile
             corel  EPS format for CorelDRAW
     <SNIP>

If the aqua terminal is not listed then you might have no chance to plot in an 
AquaTerm without recompilation of the Gnuplot sources.

Best regards

   Thomas

From jwe at octave.org  Sun Sep 27 11:10:30 2009
From: jwe at octave.org (John W. Eaton)
Date: Sun, 27 Sep 2009 12:10:30 -0400
Subject: Octave Version 3.2.3 Released
Message-ID: <19135.36470.884426.946310@segfault.lan>

GNU Octave version 3.2.3 is now available from ftp.gnu.org/gnu/octave
and ftp.octave.org/pub/octave.  The MD5 check sums are

  eecaa1a3b2da3b94fa7bcebbc4f1a1f2  octave-3.2.3.tar.bz2
  a3805ed3838c76f4973d1ef7615d5f3c  octave-3.2.3.tar.gz
  4297b33b7d5bded56c0b522729e2bb19  octave-3.2.2-3.2.3.patch.bz2
  53253104aa8201a8c64be712a5208948  octave-3.2.2-3.2.3.patch.gz

For a summary of important user-visible changes for version 3.2, see
the NEWS file (ftp://ftp.octave.org/pub/octave/NEWS-3.2), or type
"news" at the Octave prompt.

Report bugs to <bug at octave.org> (but first, please read
http://www.octave.org/bugs.html to learn how to write a helpful report).

If you find Octave useful, please support its development by contributing
to the project.  See  http://www.octave.org/help-wanted.html for more
information about how you can help.

From hellyj at ucsd.edu  Sun Sep 27 14:49:17 2009
From: hellyj at ucsd.edu (Helly John)
Date: Sun, 27 Sep 2009 12:49:17 -0700
Subject: Problem with octave/gnuplot/aquaterm on Snow Leopard
In-Reply-To: <4ABF4337.8030405@gmx.net>
References: <A59396EE-76FB-4027-887F-926E497518C0@ucsd.edu>
	<E59D6485-57DB-44F3-A837-05F1043A5192@ucsd.edu>
	<4ABF4337.8030405@gmx.net>
Message-ID: <864AA63F-44E0-48D1-90D6-15A90DFD1D29@ucsd.edu>

Thanks.  Acutally, it's related but also different.  It wasn't  
resolving against the library as you pointed out but for a obscure  
reason.  I found a discussion of the problem here:

http://macresearch.org/gnuplot-aquaterm-snow-leopard-1061

On Sep 27, 2009, at 3:49 AM, Thomas Treichl wrote:

Helly John schrieb:
> Hi.
>
> I'm new to Octave and gnuplot and Aquaterm but have just installed all
> of these on my system.  Here's the configuration:
>
> - Mac OS X 10.6.1
> - AquaTerm v1.0.1 (1.0.1)
> - Octave-3.2.2
> - GNUPLOT 4.2 patchlevel 6
>
> When I try to plot in octave this is what happens:
> *****************************************************************
> octave-3.2.2:1> Y=[1:10]
> Y =
>
>     1    2    3    4    5    6    7    8    9   10
>
> octave-3.2.2:2> plot(Y)
> Unknown or ambiguous terminal name 'aqua'
>
> gnuplot> set terminal aqua enhanced title "Figure 1" size 560 420
>                       ^
>          line 0: unknown or ambiguous terminal type; type just 'set
> terminal' for a list
>
>          line 0: No terminal defined
>
>
> gnuplot> plot "-" binary format='%float64' record=10 using ($1):($2)
> axes x1y1 title "" with lines linestyle 1 ;
>               ^
>          line 0: use 'set term' to set terminal type first
>
> octave-3.2.2:3> system ("echo $GNUTERM")
> aqua
> ans = 0
> octave-3.2.2:4>
> *****************************************************************
>
> Do you have any suggestions as to what might be wrong?
> Thanks.

Hi,

it looks like that your Gnuplot 4.2.6 has not been compiled against
libaquaterm.dylib. It's not just that you need to install AquaTerm on  
your Mac,
the one who compiles the sources that you have got needs to compile  
against
libaquaterm.dylib.
At least I hope that this is your problem. You can check this if you  
type "set
terminal" after starting Gnuplot and then check the second line for  
the "aqua"
terminal. On my system with Gnuplot 4.2.5 this looks like

   gnuplot> set terminal
   Available terminal types:
              aqua  Interface to graphics terminal server for Mac OS X
               cgm  Computer Graphics Metafile
             corel  EPS format for CorelDRAW
     <SNIP>

If the aqua terminal is not listed then you might have no chance to  
plot in an
AquaTerm without recompilation of the Gnuplot sources.

Best regards

   Thomas

-----------------------------------------------
John Helly / San Diego Supercomputer Center & Climate, Atmospheric  
Science, and Physical Oceanography/Scripps Institution of  
Oceanography / UCSD / +01 760 840 8660 mobile / stonesteps (Skype) /  
stonesteps7 (iChat) / http://www.sdsc.edu/~hellyj




From nasreddinhodja1208 at gmail.com  Tue Sep 29 01:04:18 2009
From: nasreddinhodja1208 at gmail.com (Nasreddin Hodja)
Date: Mon, 28 Sep 2009 23:04:18 -0700
Subject: Fwd: can not get octave to plot - posting again
In-Reply-To: <d52057560908232046v3967a0e5r28d5bf76df14c527@mail.gmail.com>
References: <d52057560908231451y7d513486uba2b13bad4b38cb1@mail.gmail.com>
	<Pine.LNX.4.63.0908240311450.32066@gauss.maths.ed.ac.uk>
	<d52057560908232046v3967a0e5r28d5bf76df14c527@mail.gmail.com>
Message-ID: <d52057560909282304y444f8b3dra666f3b286afea95@mail.gmail.com>

Thomas,

Below are the error messages I get when I type in the commands you
suggested.

I had not seen your response before, I wish I had... I have been trying to
plot my graphs outside of octave in the past month but it started getting
tedious. I hope to figure this plotting problem out soon.

Thank you for offering help,
--tolga


octave-3.2.2:96>  ostream = __gnuplot_open_stream__ (2, gcf ())
ostream =

   11   12

error: __gnuplot_get_var__: Can not make fifo (No such file or directory)
error: called from:
error:
/Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
at line 69, column 7
error:
/Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
at line 39, column 18
error:
/Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
at line 93, column 3
error:
/Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
at line 88, column 5
octave-3.2.2:97>
octave-3.2.2:97>  gpin_name = tmpnam ()
gpin_name = /local/tmp/oct-0C8olE
octave-3.2.2:98>
octave-3.2.2:98>  [err, msg] = mkfifo (gpin_name, 6*8*8)
err = -1
msg = No such file or directory



---------------------
Hi,

I put some Octave commands here - can you please copy and paste them into a
running Octave session and then please send back all the output of these
commands:

 ostream = __gnuplot_open_stream__ (2, gcf ())

 gpin_name = tmpnam ()

 [err, msg] = mkfifo (gpin_name, 6*8*8)

Best regards,

 Thomas



On Sun, Aug 23, 2009 at 8:46 PM, Nasreddin Hodja <
nasreddinhodja1208 at gmail.com> wrote:

>
> Thank you for your responses. Ben's suggestion actually makes an empty plot
> box pop - I did not have that before... But the plot window is blank and the
> error I get is still the same.
>
> *octave-3.2.2:1> setenv ("GNUTERM", "x11")
> octave-3.2.2:2> plot(1:10)
>
> error: __gnuplot_get_var__: Can not make fifo (No such file or directory)
> error: called from:
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
> at line 69, column 7
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
> at line 39, column 18
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
> at line 93, column 3
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
> at line 88, column 5
> *
>
>
> I tried Leo's suggestion as well...Created a .octaverc and pasted the line:
> *gnuplot_binary("/Applications/Gnuplot.app/Contents/Resources/bin/gnuplot")
>
> *in it. The behavior is the same. So I assume location of gnuplot is known
> by octave, but not really sure... When I do not type * setenv ("GNUTERM",
> "x11") *I do get a pop up blank plot window but when I do, again the error
> is:
>
> *octave-3.2.2:1>        setenv ("GNUTERM", "x11")
> octave-3.2.2:2> plot(1:10)
> error: __gnuplot_get_var__: Can not make fifo (No such file or directory)
> error: called from:
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
> at line 69, column 7
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
> at line 39, column 18
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
> at line 93, column 3
> error:
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
> at line 88, column 5
> *
> just to check at the end:*
> octave-3.2.2:3> gnuplot_binary
> ans = /Applications/Gnuplot.app/Contents/Resources/bin/gnuplot
> *
>
> Any other thoughts?
>
> Thanks,
> --tolga
>
>
>
>
>
>
>
> On Sun, Aug 23, 2009 at 7:13 PM, Leo Butler <l.butler at ed.ac.uk> wrote:
>
>>
>>
>> On Sun, 23 Aug 2009, Nasreddin Hodja wrote:
>>
>> < I hope this one goes through...
>> <
>> < Thanks in advance,
>> < --tolga
>> <
>> <
>> < ---------- Forwarded message ----------
>> < From: Nasreddin Hodja <nasreddinhodja1208 at gmail.com>
>> < Date: Sat, Aug 22, 2009 at 10:16 PM
>> < Subject: can not get octave to plot
>> < To: help-octave at octave.org
>> <
>> <
>> < Hello all,
>> <
>> < My OS: Mac OSX 10.5.6
>> < My Octave version: 3.2.2
>> < My GnuPlot version: 4.2.5
>> <
>> < I installed both GnuPlot and Octave.  Also installed control toolbox for
>> < octave, which works fine...
>> <
>> < But *when I try to use the ?plot? function in Octave I get the following
>> < error pasted below:**
>> < *
>> < FYI: I run X11 beforehand and I already set the DISPLAY to 0.0 in my
>> .zshrc.
>> < The gnuterm is set to aqua (no idea what that is but it should be ok, as
>> I
>> < followed many threads on this - though i don't think it changes anything
>> < when I set it to x11)
>> <
>> < ERROR is:
>> <
>> < octave-3.2.2:1> plot(1:10)
>> < error: __gnuplot_get_var__: Can not make fifo (No such file or
>> directory)
>>
>> I think this error indicates octave cannot find gnuplot. You can tell
>> octave the location of gnuplot by putting
>>
>> gnuplot_binary("path to gnuplot");
>>
>> in your .octaverc file in your home directory.
>>
>> Leo
>>
>> --
>> The University of Edinburgh is a charitable body, registered in
>> Scotland, with registration number SC005336.
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090928/420a1cd6/attachment.html 

From snowboardshome at gmx.de  Tue Sep 29 04:33:29 2009
From: snowboardshome at gmx.de (s4y)
Date: Tue, 29 Sep 2009 02:33:29 -0700 (PDT)
Subject: Importing data problems
Message-ID: <25659793.post@talk.nabble.com>


Hello, i can't load my file into octave.
It lokks like this:

time temperature1 temperature2
16:15:07 15,3 32,4
16:15:17 15,6 45,4
16:15:27 17,5 56,6

First of all there is a problem with the commata. Is it possible to load it
without changing the original file. Further the time value including " : "
creates an error. And i dont know how to erase the text line in the imported
file.
Thanks in advance.
Sebastian

-- 
View this message in context: http://www.nabble.com/Importing-data-problems-tp25659793p25659793.html
Sent from the Octave - General mailing list archive at Nabble.com.


From jwe at octave.org  Tue Sep 29 11:22:11 2009
From: jwe at octave.org (John W. Eaton)
Date: Tue, 29 Sep 2009 12:22:11 -0400
Subject: Importing data problems
In-Reply-To: <25659793.post@talk.nabble.com>
References: <25659793.post@talk.nabble.com>
Message-ID: <19138.13363.393013.244615@segfault.lan>

On 29-Sep-2009, s4y wrote:

| Hello, i can't load my file into octave.
| It lokks like this:
| 
| time temperature1 temperature2
| 16:15:07 15,3 32,4
| 16:15:17 15,6 45,4
| 16:15:27 17,5 56,6
| 
| First of all there is a problem with the commata. Is it possible to load it
| without changing the original file. Further the time value including " : "
| creates an error. And i dont know how to erase the text line in the imported
| file.

Read it with fopen and some combination of fscanf or fgetl+conversion
to whatever data types you need.

Or, take a look at the textread function from the Octave Forge io
package.

jwe

From ar63 at evansville.edu  Tue Sep 29 11:19:44 2009
From: ar63 at evansville.edu (Richardson, Anthony)
Date: Tue, 29 Sep 2009 11:19:44 -0500
Subject: Importing data problems
In-Reply-To: <25659793.post@talk.nabble.com>
References: <25659793.post@talk.nabble.com>
Message-ID: <2318D9796D5B9D41B85612B5C87F2240025034E0@UEEXCHANGE.evansville.edu>

Here's my attempt:

fid = fopen("temp.dat","r");
%Read header
h = fgetl(fid);
index = 1;
t1 = [];
t2 = [];
time = ["        "];
while(1)
  [t n1 n2 n3 n4 count] = fscanf(fid,"%s %d,%d %d,%d" ,"C");
  time(index,:) = t;
  t1(index) = [n1+n2/10];
  t2(index) = [n3+n4/10];
  index++;
  if (feof(fid)) break; endif
endwhile
fclose(fid);


> -----Original Message-----
> From: s4y [mailto:snowboardshome at gmx.de]
> Sent: Tuesday, September 29, 2009 4:33 AM
> To: help-octave at octave.org
> Subject: Importing data problems
> 
> 
> Hello, i can't load my file into octave.
> It lokks like this:
> 
> time temperature1 temperature2
> 16:15:07 15,3 32,4
> 16:15:17 15,6 45,4
> 16:15:27 17,5 56,6
> 
> First of all there is a problem with the commata. Is it possible to
> load it
> without changing the original file. Further the time value including "
> : "
> creates an error. And i dont know how to erase the text line in the
> imported
> file.
> Thanks in advance.
> Sebastian
> 
> --
> View this message in context: http://www.nabble.com/Importing-data-
> problems-tp25659793p25659793.html
> Sent from the Octave - General mailing list archive at Nabble.com.
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave


From dcottingham00 at comcast.net  Tue Sep 29 12:28:05 2009
From: dcottingham00 at comcast.net (dcottingham00 at comcast.net)
Date: Tue, 29 Sep 2009 17:28:05 +0000 (UTC)
Subject: another "gset term" replacement question
Message-ID: <899453044.176741254245285627.JavaMail.root@sz0019a.westchester.pa.mail.comcast.net>

I often have a need to run octave scripts in the background, sometimes scripts that make plots. With octave 2.x, I would preceed any such plotting with "gset term postscript". 

Now that gset is gone, I do the plotting and the use "print -dpsc" to put it into a postscript file. But if I try to run such a script in background, it hangs when it gets to the plotting. 

How can I direct the script to make the plot postscript in the first place? Or is there any other way to stop it from hanging? Is there any way to control where octave's invocation of gnuplot tries to put the plot? 

Thanks, 
Dave Cottingham 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090929/2540560b/attachment.html 

From lindnerben at gmx.net  Tue Sep 29 13:31:43 2009
From: lindnerben at gmx.net (Benjamin Lindner)
Date: Tue, 29 Sep 2009 20:31:43 +0200
Subject: startup with .octaverc on Windows Vista
In-Reply-To: <20090924061926.GA16446@simba>
References: <20090921172000.GA8578@simba>	<2318D9796D5B9D41B85612B5C87F224002350E5A@UEEXCHANGE.evansville.edu>
	<20090924061926.GA16446@simba>
Message-ID: <4AC2528F.3000604@gmx.net>

Bart Vandewoestyne wrote:
> On Mon, Sep 21, 2009 at 01:43:26PM -0500, Richardson, Anthony wrote:
>> We have a similar setup.   We set HOME in the
>> C:/octave/3.22.2_gcc-4.3.0/share/octave/site/m/startup/octaverc
>> file, with the following lines:
>>
>> setenv('HOME','h:\');
>> cd ~;
>>
>> This allows the octave HOME env variable to be set differently 
>> than a Vista global HOME environment variable.  An .octaverc 
>> file in h:\ is still executed.  Setting HOME from Windows
>> should work too, but we have other software that uses the HOME
>> environment variable and requires that it be set to something
>> else.
>>
>> Tony Richardson
> 
> Thanks for your reply, Tony.  In our computer-rooms, we are doing
> it now the way you describe, and everything works nicely.  The
> Octave desktop-icon is now shining on all our student-computers
> :-)
> 
> I do have another question related to the installer though...
> 
> For our sysadmins, it would be comfortable if they can do a
> silent (unattended) install.  For as far as I can see, Octave
> uses the NSIS installer system (http://nsis.sourceforge.net) and
> it has an option /S with which one can do a silent install (see
> for example http://unattended.sourceforge.net/installers.php)
> 
> However, if one performs a silent install, then apparently the
> installation is done with the *default* options, that is, no
> Octave-Forge packages are installed and one cannot select another
> Atlas library.
> 
> Is it somehow possible to do a silent (unattended) install *with*
> specifying that Octave-Forge should also be installed?  Does the
> NSIS installer system allow a user to specify certain
> installation options somehow?

This is indeed a good question.
I did consult the NSIS manual, and it seems that there is no supoprt for 
an answer file for unattended installs, which means it would need to be 
programmed by hand.
I am not an expert on nsis, and frankly I am not overly eager to start 
coding this by hand.

One possibility that you can try is to install octave once in the 
configuration desired, and then simply copy the entire directory tree 
as-is to a different computer. Octave itself is relocatable, and it is 
portable, so there is no need to install it to get a running version.
What's not fully relocatable at the moment is the octave-forge packages 
(the package manager keeps track of absolute installation paths).
So in order to get a copied directory tree working you need to start 
octave and execute "pkg rebuild".
You can omit this step if you copy octave into the same path on a remote 
computer (since then no paths actually have changed, it's just a 
different machine).

What you cannot do this way however is to get different ATLAS architectures.

benjamin

From Thomas.Treichl at gmx.net  Tue Sep 29 13:40:47 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Tue, 29 Sep 2009 20:40:47 +0200
Subject: Fwd: can not get octave to plot - posting again
In-Reply-To: <d52057560909282304y444f8b3dra666f3b286afea95@mail.gmail.com>
References: <d52057560908231451y7d513486uba2b13bad4b38cb1@mail.gmail.com>	
	<Pine.LNX.4.63.0908240311450.32066@gauss.maths.ed.ac.uk>	
	<d52057560908232046v3967a0e5r28d5bf76df14c527@mail.gmail.com>
	<d52057560909282304y444f8b3dra666f3b286afea95@mail.gmail.com>
Message-ID: <4AC254AF.2070604@gmx.net>

Nasreddin Hodja schrieb:
> 
> Thomas,
> 
> Below are the error messages I get when I type in the commands you 
> suggested.
> 
> I had not seen your response before, I wish I had... I have been trying 
> to plot my graphs outside of octave in the past month but it started 
> getting tedious. I hope to figure this plotting problem out soon.
> 
> Thank you for offering help,
> --tolga
> 
> 
> octave-3.2.2:96>  ostream = __gnuplot_open_stream__ (2, gcf ())
> ostream =
> 
>    11   12
> 
> error: __gnuplot_get_var__: Can not make fifo (No such file or directory)
> error: called from:
> error:   
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m 
> at line 69, column 7
> error:   
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m 
> at line 39, column 18
> error:   
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m 
> at line 93, column 3
> error:   
> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m 
> at line 88, column 5
> octave-3.2.2:97>
> octave-3.2.2:97>  gpin_name = tmpnam ()
> gpin_name = /local/tmp/oct-0C8olE
> octave-3.2.2:98>
> octave-3.2.2:98>  [err, msg] = mkfifo (gpin_name, 6*8*8)
> err = -1
> msg = No such file or directory
> 
> 
> 
> ---------------------
> Hi,
> 
> I put some Octave commands here - can you please copy and paste them 
> into a running Octave session and then please send back all the output 
> of these commands:
> 
>  ostream = __gnuplot_open_stream__ (2, gcf ())
> 
>  gpin_name = tmpnam ()
> 
>  [err, msg] = mkfifo (gpin_name, 6*8*8)
> 
> Best regards,
> 
>  Thomas
> 
> 
> 
> On Sun, Aug 23, 2009 at 8:46 PM, Nasreddin Hodja 
> <nasreddinhodja1208 at gmail.com <mailto:nasreddinhodja1208 at gmail.com>> wrote:
> 
> 
>     Thank you for your responses. Ben's suggestion actually makes an
>     empty plot box pop - I did not have that before... But the plot
>     window is blank and the error I get is still the same.
> 
>     */octave-3.2.2:1> setenv ("GNUTERM", "x11")
>     octave-3.2.2:2> plot(1:10)
> 
> 
>     error: __gnuplot_get_var__: Can not make fifo (No such file or
>     directory)
>     error: called from:
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
>     at line 69, column 7
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
>     at line 39, column 18
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
>     at line 93, column 3
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
>     at line 88, column 5
>     /*
> 
> 
>     I tried Leo's suggestion as well...Created a .octaverc and pasted
>     the line:
>     */gnuplot_binary("/Applications/Gnuplot.app/Contents/Resources/bin/gnuplot")
> 
>     /*in it. The behavior is the same. So I assume location of gnuplot
>     is known by octave, but not really sure... When I do not type */
>     setenv ("GNUTERM", "x11") /*I do get a pop up blank plot window but
>     when I do, again the error is:
> 
>     /*octave-3.2.2:1>        setenv ("GNUTERM", "x11")
>     octave-3.2.2:2> plot(1:10)
> 
>     error: __gnuplot_get_var__: Can not make fifo (No such file or
>     directory)
>     error: called from:
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
>     at line 69, column 7
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
>     at line 39, column 18
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
>     at line 93, column 3
>     error:  
>     /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
>     at line 88, column 5
>     */
>     just to check at the end:/*
>     octave-3.2.2:3> gnuplot_binary
>     ans = /Applications/Gnuplot.app/Contents/Resources/bin/gnuplot
>     */
> 
>     Any other thoughts?
> 
>     Thanks,
>     --tolga
> 
> 
> 
> 
> 
> 
> 
>     On Sun, Aug 23, 2009 at 7:13 PM, Leo Butler <l.butler at ed.ac.uk
>     <mailto:l.butler at ed.ac.uk>> wrote:
> 
> 
> 
>         On Sun, 23 Aug 2009, Nasreddin Hodja wrote:
> 
>         < I hope this one goes through...
>         <
>         < Thanks in advance,
>         < --tolga
>         <
>         <
>         < ---------- Forwarded message ----------
>         < From: Nasreddin Hodja <nasreddinhodja1208 at gmail.com
>         <mailto:nasreddinhodja1208 at gmail.com>>
>         < Date: Sat, Aug 22, 2009 at 10:16 PM
>         < Subject: can not get octave to plot
>         < To: help-octave at octave.org <mailto:help-octave at octave.org>
>         <
>         <
>         < Hello all,
>         <
>         < My OS: Mac OSX 10.5.6
>         < My Octave version: 3.2.2
>         < My GnuPlot version: 4.2.5
>         <
>         < I installed both GnuPlot and Octave.  Also installed control
>         toolbox for
>         < octave, which works fine...
>         <
>         < But *when I try to use the ?plot? function in Octave I get the
>         following
>         < error pasted below:**
>         < *
>         < FYI: I run X11 beforehand and I already set the DISPLAY to 0.0
>         in my .zshrc.
>         < The gnuterm is set to aqua (no idea what that is but it should
>         be ok, as I
>         < followed many threads on this - though i don't think it
>         changes anything
>         < when I set it to x11)
>         <
>         < ERROR is:
>         <
>         < octave-3.2.2:1> plot(1:10)
>         < error: __gnuplot_get_var__: Can not make fifo (No such file or
>         directory)
> 
>         I think this error indicates octave cannot find gnuplot. You can
>         tell
>         octave the location of gnuplot by putting
> 
>         gnuplot_binary("path to gnuplot");
> 
>         in your .octaverc file in your home directory.
> 
>         Leo
> 
>         --
>         The University of Edinburgh is a charitable body, registered in
>         Scotland, with registration number SC005336.

Sorry, I must have overlooked your answer. Ok, let's see what we can do now. The 
problem you have has nothing to do with where the binary of gnuplot is located. 
Your problem is that you cannot create a fifo special file on your Mac.

I think the problem is somewhere with your tmpnam() function because it returns 
a result to /local/tmp. On my system a call to tmpnam() within Octave gives the 
result

   octave-3.2.0:1> tmpnam ()
   ans = /var/tmp/oct-KmvZyU

First question now is, is there a path on your system which is called 
/local/tmp? Open a Terminal.app and check this

   ls -l /local/tmp

I expect this path does not exist on your Mac or it has been created manually?! 
The next question is, is there a global /tmp directory on your Mac, check

   ls -l /tmp

and also please type this line in your Terminal.app

   echo $TMPDIR

I also expect that your Octave mkfifo() command works correctly, can you please 
check in Octave.app

   [verr, vmsg] = mkfifo ("test.txt", 666)
   delete ("test.txt")

Meanwhile best regards

   Thomas

From benjamin.abbott at tqs.com  Tue Sep 29 14:58:52 2009
From: benjamin.abbott at tqs.com (Ben Abbott)
Date: Tue, 29 Sep 2009 15:58:52 -0400
Subject: another "gset term" replacement question
In-Reply-To: <109821932292442390389609557899284330578-Webmail@me.com>
References: <899453044.176741254245285627.JavaMail.root@sz0019a.westchester.pa.mail.comcast.net>
	<109821932292442390389609557899284330578-Webmail@me.com>
Message-ID: <1408EA1E-C2B6-4CC6-B87C-B5A73CC2E71C@tqs.com>

>  On Tuesday, September 29, 2009, at 01:28PM, <dcottingham00 at comcast.net 
> > wrote:
>
> I often have a need to run octave scripts in the background,  
> sometimes scripts that make plots. With octave 2.x, I would preceed  
> any such plotting with "gset term postscript".
>
> Now that gset is gone, I do the plotting and the use "print -dpsc"  
> to put it into a postscript file. But if I try to run such a script  
> in background, it hangs when it gets to the plotting.
>
> How can I direct the script to make the plot postscript in the first  
> place? Or is there any other way to stop it from hanging? Is there  
> any way to control where octave's invocation of gnuplot tries to put  
> the plot?
>
> Thanks,
> Dave Cottingham

To produce a postscript output ...

	print -dpsc file.ps

To suspend generation of figures on screen, you can try ...

	close all
	set (0, "defaultfigurevisible", "off")

To continue with plots to the screen ...

	close all
	set (0, "defaultfigurevisible", "on")

Ben




From dcottingham00 at comcast.net  Tue Sep 29 20:39:59 2009
From: dcottingham00 at comcast.net (dcottingham00 at comcast.net)
Date: Wed, 30 Sep 2009 01:39:59 +0000 (UTC)
Subject: another "gset term" replacement question
In-Reply-To: <1408EA1E-C2B6-4CC6-B87C-B5A73CC2E71C@tqs.com>
Message-ID: <731821141.377071254274799478.JavaMail.root@sz0019a.westchester.pa.mail.comcast.net>

Thanks. Your suggestion does indeed stop the plot from appearing on the screen -- but when run in the background, the process still stops. 

BTW, I shouldn't have said it hangs if run in the background -- it goes into the "stopped" state. And further experiments have revealed that a script that does plotting, but does not do "print", does not go into "stopped". Even if I omit the part that sets visible off. It only stops if I include a "print." 

Maybe I should write a bug report -- I can't think of any good reason why "print" should have to run in the foreground. 

- Dave Cottingham 

----- Original Message ----- 
From: "Ben Abbott" <benjamin.abbott at tqs.com> 
To: dcottingham00 at comcast.net 
Cc: help-octave at octave.org 
Sent: Tuesday, September 29, 2009 3:58:52 PM GMT -05:00 US/Canada Eastern 
Subject: Re: another "gset term" replacement question 

> On Tuesday, September 29, 2009, at 01:28PM, <dcottingham00 at comcast.net 
> > wrote: 
> 
> I often have a need to run octave scripts in the background, 
> sometimes scripts that make plots. With octave 2.x, I would preceed 
> any such plotting with "gset term postscript". 
> 
> Now that gset is gone, I do the plotting and the use "print -dpsc" 
> to put it into a postscript file. But if I try to run such a script 
> in background, it hangs when it gets to the plotting. 
> 
> How can I direct the script to make the plot postscript in the first 
> place? Or is there any other way to stop it from hanging? Is there 
> any way to control where octave's invocation of gnuplot tries to put 
> the plot? 
> 
> Thanks, 
> Dave Cottingham 

To produce a postscript output ... 

print -dpsc file.ps 

To suspend generation of figures on screen, you can try ... 

close all 
set (0, "defaultfigurevisible", "off") 

To continue with plots to the screen ... 

close all 
set (0, "defaultfigurevisible", "on") 

Ben 



-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090930/25448040/attachment.html 

From nasreddinhodja1208 at gmail.com  Wed Sep 30 01:43:00 2009
From: nasreddinhodja1208 at gmail.com (Nasreddin Hodja)
Date: Tue, 29 Sep 2009 23:43:00 -0700
Subject: Fwd: can not get octave to plot - posting again
In-Reply-To: <4AC254AF.2070604@gmx.net>
References: <d52057560908231451y7d513486uba2b13bad4b38cb1@mail.gmail.com>
	<Pine.LNX.4.63.0908240311450.32066@gauss.maths.ed.ac.uk>
	<d52057560908232046v3967a0e5r28d5bf76df14c527@mail.gmail.com>
	<d52057560909282304y444f8b3dra666f3b286afea95@mail.gmail.com>
	<4AC254AF.2070604@gmx.net>
Message-ID: <d52057560909292343q146083f6o4001463c3156ae93@mail.gmail.com>

On Tue, Sep 29, 2009 at 11:40 AM, Thomas Treichl <Thomas.Treichl at gmx.net>wrote:

> Nasreddin Hodja schrieb:
>
>>
>> Thomas,
>>
>> Below are the error messages I get when I type in the commands you
>> suggested.
>>
>> I had not seen your response before, I wish I had... I have been trying to
>> plot my graphs outside of octave in the past month but it started getting
>> tedious. I hope to figure this plotting problem out soon.
>>
>> Thank you for offering help,
>> --tolga
>>
>>
>> octave-3.2.2:96>  ostream = __gnuplot_open_stream__ (2, gcf ())
>> ostream =
>>
>>   11   12
>>
>> error: __gnuplot_get_var__: Can not make fifo (No such file or directory)
>> error: called from:
>> error:
>> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
>> at line 69, column 7
>> error:
>> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
>> at line 39, column 18
>> error:
>> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
>> at line 93, column 3
>> error:
>> /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
>> at line 88, column 5
>> octave-3.2.2:97>
>> octave-3.2.2:97>  gpin_name = tmpnam ()
>> gpin_name = /local/tmp/oct-0C8olE
>> octave-3.2.2:98>
>> octave-3.2.2:98>  [err, msg] = mkfifo (gpin_name, 6*8*8)
>> err = -1
>> msg = No such file or directory
>>
>>
>>
>> ---------------------
>> Hi,
>>
>> I put some Octave commands here - can you please copy and paste them into
>> a running Octave session and then please send back all the output of these
>> commands:
>>
>>  ostream = __gnuplot_open_stream__ (2, gcf ())
>>
>>  gpin_name = tmpnam ()
>>
>>  [err, msg] = mkfifo (gpin_name, 6*8*8)
>>
>> Best regards,
>>
>>  Thomas
>>
>>
>>
>> On Sun, Aug 23, 2009 at 8:46 PM, Nasreddin Hodja <
>> nasreddinhodja1208 at gmail.com <mailto:nasreddinhodja1208 at gmail.com>>
>> wrote:
>>
>>
>>    Thank you for your responses. Ben's suggestion actually makes an
>>    empty plot box pop - I did not have that before... But the plot
>>    window is blank and the error I get is still the same.
>>
>>    */octave-3.2.2:1> setenv ("GNUTERM", "x11")
>>    octave-3.2.2:2> plot(1:10)
>>
>>
>>    error: __gnuplot_get_var__: Can not make fifo (No such file or
>>    directory)
>>    error: called from:
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
>>    at line 69, column 7
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
>>    at line 39, column 18
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
>>    at line 93, column 3
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
>>    at line 88, column 5
>>    /*
>>
>>
>>    I tried Leo's suggestion as well...Created a .octaverc and pasted
>>    the line:
>>
>>  */gnuplot_binary("/Applications/Gnuplot.app/Contents/Resources/bin/gnuplot")
>>
>>    /*in it. The behavior is the same. So I assume location of gnuplot
>>    is known by octave, but not really sure... When I do not type */
>>    setenv ("GNUTERM", "x11") /*I do get a pop up blank plot window but
>>    when I do, again the error is:
>>
>>    /*octave-3.2.2:1>        setenv ("GNUTERM", "x11")
>>    octave-3.2.2:2> plot(1:10)
>>
>>    error: __gnuplot_get_var__: Can not make fifo (No such file or
>>    directory)
>>    error: called from:
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__gnuplot_get_var__.m
>>    at line 69, column 7
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_axes__.m
>>    at line 39, column 18
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/__go_draw_figure__.m
>>    at line 93, column 3
>>    error:
>>  /Applications/Octave.app/Contents/Resources/share/octave/3.2.2/m/plot/gnuplot_drawnow.m
>>    at line 88, column 5
>>    */
>>    just to check at the end:/*
>>    octave-3.2.2:3> gnuplot_binary
>>    ans = /Applications/Gnuplot.app/Contents/Resources/bin/gnuplot
>>    */
>>
>>    Any other thoughts?
>>
>>    Thanks,
>>    --tolga
>>
>>
>>
>>
>>
>>
>>
>>    On Sun, Aug 23, 2009 at 7:13 PM, Leo Butler <l.butler at ed.ac.uk
>>    <mailto:l.butler at ed.ac.uk>> wrote:
>>
>>
>>
>>        On Sun, 23 Aug 2009, Nasreddin Hodja wrote:
>>
>>        < I hope this one goes through...
>>        <
>>        < Thanks in advance,
>>        < --tolga
>>        <
>>        <
>>        < ---------- Forwarded message ----------
>>        < From: Nasreddin Hodja <nasreddinhodja1208 at gmail.com
>>        <mailto:nasreddinhodja1208 at gmail.com>>
>>        < Date: Sat, Aug 22, 2009 at 10:16 PM
>>        < Subject: can not get octave to plot
>>        < To: help-octave at octave.org <mailto:help-octave at octave.org>
>>        <
>>        <
>>        < Hello all,
>>        <
>>        < My OS: Mac OSX 10.5.6
>>        < My Octave version: 3.2.2
>>        < My GnuPlot version: 4.2.5
>>        <
>>        < I installed both GnuPlot and Octave.  Also installed control
>>        toolbox for
>>        < octave, which works fine...
>>        <
>>        < But *when I try to use the ?plot? function in Octave I get the
>>        following
>>        < error pasted below:**
>>        < *
>>        < FYI: I run X11 beforehand and I already set the DISPLAY to 0.0
>>        in my .zshrc.
>>        < The gnuterm is set to aqua (no idea what that is but it should
>>        be ok, as I
>>        < followed many threads on this - though i don't think it
>>        changes anything
>>        < when I set it to x11)
>>        <
>>        < ERROR is:
>>        <
>>        < octave-3.2.2:1> plot(1:10)
>>        < error: __gnuplot_get_var__: Can not make fifo (No such file or
>>        directory)
>>
>>        I think this error indicates octave cannot find gnuplot. You can
>>        tell
>>        octave the location of gnuplot by putting
>>
>>        gnuplot_binary("path to gnuplot");
>>
>>        in your .octaverc file in your home directory.
>>
>>        Leo
>>
>>        --
>>        The University of Edinburgh is a charitable body, registered in
>>        Scotland, with registration number SC005336.
>>
>
> Sorry, I must have overlooked your answer. Ok, let's see what we can do
> now. The problem you have has nothing to do with where the binary of gnuplot
> is located. Your problem is that you cannot create a fifo special file on
> your Mac.
>
> I think the problem is somewhere with your tmpnam() function because it
> returns a result to /local/tmp. On my system a call to tmpnam() within
> Octave gives the result
>
>  octave-3.2.0:1> tmpnam ()
>  ans = /var/tmp/oct-KmvZyU
>
> First question now is, is there a path on your system which is called
> /local/tmp? Open a Terminal.app and check this
>
>  ls -l /local/tmp
>
> I expect this path does not exist on your Mac or it has been created
> manually?! The next question is, is there a global /tmp directory on your
> Mac, check
>
>  ls -l /tmp
>
> and also please type this line in your Terminal.app
>
>  echo $TMPDIR
>
> I also expect that your Octave mkfifo() command works correctly, can you
> please check in Octave.app
>
>  [verr, vmsg] = mkfifo ("test.txt", 666)
>  delete ("test.txt")
>
> Meanwhile best regards
>
>  Thomas
>



Thomas,

The results are interesting. Do I need to reset TMPDIR variable to something
else. It seems to point to something non-existent...or just create a
/local/tmp?

*ls -l /local/tmp
ls: /local/tmp: No such file or directory

ls -l /tmp
lrwxr-xr-x@ 1 root  admin  12 Jun  4 10:15 /tmp@ -> /private/tmp

echo $TMPDIR
*
*/local/tmp

octave-3.2.2:368> tmpnam ()
ans = /local/tmp/oct-a1D8LQ

octave-3.2.2:366>  [verr, vmsg] = mkfifo ("test.txt", 666)
verr = 0
vmsg =
octave-3.2.2:367>  delete ("test.txt")
*
Thanks and best regards...

--tolga
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090929/d8d7eb74/attachment-0001.html 

From eric.salemi at septentrio.com  Wed Sep 30 04:49:51 2009
From: eric.salemi at septentrio.com (Eric Salemi)
Date: Wed, 30 Sep 2009 11:49:51 +0200
Subject: Error on a simple struct assignment
Message-ID: <4AC329BF.9070204@septentrio.com>

Hi,

I'm working with Octave 3.2.2 on a windows XP machine. I type the
following 3 commands and get a error after the 3rd (you'll notice the
2nd and 3rd commands are exactly the same):

 > s = struct;

 > s(1,1,1).field1 = 0
s =
{
   field1 = 0
}

 > s(1,1,1).field1 = 0
error: invalid assignment to cs-list outside multiple assignment.
error: assignment failed, or no method for `struct = scalar'

Maybe there is something I don't understand, but I don't see why Octave
consider s(1,1,1).field1 as a cs-list in the 3rd command. As a side
note, doing the same with (2,2,2) as index works flawlessly. Can I
consider this issue a bug? Is there a workaround?

Eric.




Septentrio has moved offices.
As of Monday 31 August 2009, Septentrio headquarters are located at:
Greenhill Campus
Interleuvenlaan 15G
3001 Leuven
Belgium.
Phone and fax numbers remain unchanged


This e-mail communication contains information that is confidential and may also be privileged. It is intended for the exclusive use of the addressees. If you are not the person or organization to whom it is addressed, you must not copy, distribute or take any action in reliance upon it. If you received this communication in error, please notify Septentrio nv immediately [ telephone +32 [0] 16 300800 ]. Septentrio nv will not accept liability for contractual commitments made by individuals employed by this company outside the scope of our business.


From snowboardshome at gmx.de  Wed Sep 30 04:32:12 2009
From: snowboardshome at gmx.de (s4y)
Date: Wed, 30 Sep 2009 02:32:12 -0700 (PDT)
Subject: Importing data problems
In-Reply-To: <2318D9796D5B9D41B85612B5C87F2240025034E0@UEEXCHANGE.evansville.edu>
References: <25659793.post@talk.nabble.com>
	<2318D9796D5B9D41B85612B5C87F2240025034E0@UEEXCHANGE.evansville.edu>
Message-ID: <25677802.post@talk.nabble.com>


Hey, thanks for your attempt. It helped alot.
I got my programme almost workin so far. But i have one problem.
My programme:
[fid] = fopen("test2.txt", "r", "native");
kopfzeile = fgetl(fid);
[m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f",[2,9]);
fclose(fid);
m 

the file test2.txt:
Text
16:09:06   32.41   32.63   36.43    36.63   36.65   34.56
16:09:13   32.39   35.63   46.43    66.83   26.65   34.56

Unfortunately he saves the matrix m this way:
 16      06        32.63    36.63   34.56    09   32.39  
 09     32.41    36.43    36.65    16        13   35.63  

he starts to write it in the first column and then he is moving on to next
column
but i need it this way:
16  09  06   32.41   32.63   36.43    36.63   36.65   34.56
16  09  13   32.39   35.63   46.43    66.83   26.65   34.56 
like in the file, line by line
But i dont know how to change this.
Hopefully someone knows a way.
Thanks in advance.

-- 
View this message in context: http://www.nabble.com/Importing-data-problems-tp25659793p25677802.html
Sent from the Octave - General mailing list archive at Nabble.com.


From Bart.Vandewoestyne at telenet.be  Wed Sep 30 08:30:11 2009
From: Bart.Vandewoestyne at telenet.be (Bart Vandewoestyne)
Date: Wed, 30 Sep 2009 15:30:11 +0200
Subject: Help-octave Digest, Vol 42, Issue 55
In-Reply-To: <mailman.2819.1254292992.21430.help-octave@octave.org>
References: <mailman.2819.1254292992.21430.help-octave@octave.org>
Message-ID: <20090930133011.GA14304@simba>

On Wed, Sep 30, 2009 at 01:43:12AM -0500, help-octave-request at octave.org wrote:
>
> This is indeed a good question.
> I did consult the NSIS manual, and it seems that there is no supoprt for 
> an answer file for unattended installs, which means it would need to be 
> programmed by hand.
> I am not an expert on nsis, and frankly I am not overly eager to start 
> coding this by hand.

I have never used NSIS in my life, but at the bottom of

http://nsis.sourceforge.net/Docs/Chapter4.html#4.12

it is mentioned:

"If your installer/uninstaller requires some more information
 that can not be gathered when silent, you can allow the user to
 specify that information on the command line and process it in
 .onInit."

and completely at the bottom is written:

"If your installer/uninstaller requires a lot of information and
 you want it to be able to be silent, you should allow the user to
 pass on a path to an answers file. This would be much more
 comfortable than writing all of the information on the command
 line."

So i guess this means that there *are* ways to do it, no?

If there are few options:
  -> use the .onInit. processing technique

If there are a lot of options:
  -> let the user pass a path to an answer file

But again... I don't know anything about NSIS and absolutely do
not have the time to dig into this... so I'm afraid this is all I
can help.

Maybe other installer systems could also be considered?  The free and
Open Source ones that I know of, are listed on my webpage
http://osswin.sourceforge.net/ (search for 'Installer systems').
I think Inno Setup is the most known, but I don't know if it is
better/worse than NSIS.

Regards,
Bart

-- 
	"Share what you know.  Learn what you don't."

From martin at mhelm.de  Wed Sep 30 11:13:33 2009
From: martin at mhelm.de (martin_helm)
Date: Wed, 30 Sep 2009 09:13:33 -0700 (PDT)
Subject: Importing data problems
In-Reply-To: <25677802.post@talk.nabble.com>
References: <25659793.post@talk.nabble.com>
	<2318D9796D5B9D41B85612B5C87F2240025034E0@UEEXCHANGE.evansville.edu>
	<25677802.post@talk.nabble.com>
Message-ID: <25684266.post@talk.nabble.com>




s4y wrote:
> 
> Hey, thanks for your attempt. It helped alot.
> I got my programme almost workin so far. But i have one problem.
> My programme:
> [fid] = fopen("test2.txt", "r", "native");
> kopfzeile = fgetl(fid);
> [m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f",[2,9]);
> fclose(fid);
> m 
> 
> the file test2.txt:
> Text
> 16:09:06   32.41   32.63   36.43    36.63   36.65   34.56
> 16:09:13   32.39   35.63   46.43    66.83   26.65   34.56
> 
> Unfortunately he saves the matrix m this way:
>  16      06        32.63    36.63   34.56    09   32.39  
>  09     32.41    36.43    36.65    16        13   35.63  
> 
> he starts to write it in the first column and then he is moving on to next
> column
> but i need it this way:
> 16  09  06   32.41   32.63   36.43    36.63   36.65   34.56
> 16  09  13   32.39   35.63   46.43    66.83   26.65   34.56 
> like in the file, line by line
> But i dont know how to change this.
> Hopefully someone knows a way.
> Thanks in advance.
> 
> 

You can use 

[fid] = fopen("test2.txt", "r", "native");
kopfzeile = fgetl(fid);
[m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f",[9,2]);
m=m';
fclose(fid);

But it has still the drawback that you have to know the number of lines in
advance.

Instead you can do something like

[fid] = fopen("test2.txt", "r", "native");
kopfzeile = fgetl(fid);
[m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f");
fclose(fid);
m = reshape(m, 9, count/9)';


Regards
Martin

-- 
View this message in context: http://www.nabble.com/Importing-data-problems-tp25659793p25684266.html
Sent from the Octave - General mailing list archive at Nabble.com.


From przemek at nist.gov  Wed Sep 30 13:11:01 2009
From: przemek at nist.gov (Przemek Klosowski)
Date: Wed, 30 Sep 2009 14:11:01 -0400
Subject: Importing data problems
In-Reply-To: <25677802.post@talk.nabble.com>
References: <25659793.post@talk.nabble.com>	<2318D9796D5B9D41B85612B5C87F2240025034E0@UEEXCHANGE.evansville.edu>
	<25677802.post@talk.nabble.com>
Message-ID: <4AC39F35.8010504@nist.gov>

On 09/30/2009 05:32 AM, s4y asked why the code:
> [fid] = fopen("test2.txt", "r", "native");
> [m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f",[2,9]);
> fclose(fid);

incorrectly chops up the following data file

16:09:06   32.41   32.63   36.43    36.63   36.65   34.56
16:09:13   32.39   35.63   46.43    66.83   26.65   34.56

The best way to think about it is that fscanf() reads a stream of data 
into a single column, which can then be reshaped. You ask for a [2,9] 
form, which chops 2-row-long columnar pieces and assembles them into 9 
columns. If you do

[m, count] = fscanf(fid, "%f:%f:%f %f %f %f %f %f %f",[9,2]);

and transpose ( m' ) you will get your desired result.

From mikelmckinney at gmail.com  Wed Sep 30 13:42:40 2009
From: mikelmckinney at gmail.com (michael mckinney)
Date: Wed, 30 Sep 2009 13:42:40 -0500
Subject: data set
Message-ID: <e164838f0909301142n66dc6a58s2a6765a5e202bd62@mail.gmail.com>

I have some data, taken in two different formats, which corresponds to
a trail- I'd like to convert the xls data I have to ascII data, or
something that would be suitable for an elevation and graphing format.
 Do you have any suggestions?  thank you, Mike McKinney.

From Thomas.Treichl at gmx.net  Wed Sep 30 13:50:34 2009
From: Thomas.Treichl at gmx.net (Thomas Treichl)
Date: Wed, 30 Sep 2009 20:50:34 +0200
Subject: Fwd: can not get octave to plot - posting again
In-Reply-To: <d52057560909292343q146083f6o4001463c3156ae93@mail.gmail.com>
References: <d52057560908231451y7d513486uba2b13bad4b38cb1@mail.gmail.com>	
	<Pine.LNX.4.63.0908240311450.32066@gauss.maths.ed.ac.uk>	
	<d52057560908232046v3967a0e5r28d5bf76df14c527@mail.gmail.com>	
	<d52057560909282304y444f8b3dra666f3b286afea95@mail.gmail.com>	
	<4AC254AF.2070604@gmx.net>
	<d52057560909292343q146083f6o4001463c3156ae93@mail.gmail.com>
Message-ID: <4AC3A87A.5060607@gmx.net>

Nasreddin Hodja schrieb:
> Thomas,
> 
> The results are interesting. Do I need to reset TMPDIR variable to 
> something else. It seems to point to something non-existent...or just 
> create a /local/tmp?
> *ls -l /local/tmp
> ls: /local/tmp: No such file or directory
> 
> ls -l /tmp
> lrwxr-xr-x@ 1 root  admin  12 Jun  4 10:15 /tmp@ -> /private/tmp
> 
> echo $TMPDIR
> *
> */local/tmp
> 
> octave-3.2.2:368> tmpnam ()
> ans = /local/tmp/oct-a1D8LQ
> 
> octave-3.2.2:366>  [verr, vmsg] = mkfifo ("test.txt", 666)
> verr = 0
> vmsg =
> octave-3.2.2:367>  delete ("test.txt")
> *
> Thanks and best regards...
> 
> --tolga

Hi again,

don't create the /local/tmp directory because there is no need for such a 
directory on a Mac. The interesting part for me is, where on your Mac is TMPDIR 
defined? Is it a local definition somewhere in your ~/.profile or ~/.bashrc file 
or is it even worst defined globally? You can quickly check in Terminal.app

   cat ~/.profile
   cat ~/.bashrc

If you have configured another user on your machine then you can check if TMPDIR 
is there globally. Maybe it's defined like this

   http://developer.apple.com/mac/library/qa/qa2001/qa1067.html

My suggestion simply is to find the TMPDIR variable definition on your system 
and remove it. There really is no need to redirect the tmp directory to such a 
stupid path. If you cannot find the definition or you want to keep it because of 
a special reason then this is a workaround for your Octave to plot in an aqua 
window that should work

   setenv ("TMPDIR", "")
   plot (1:10)

Best regards

   Thomas

From ct529 at york.ac.uk  Wed Sep 30 15:11:50 2009
From: ct529 at york.ac.uk (Corrado)
Date: Wed, 30 Sep 2009 21:11:50 +0100
Subject: A point in a vector?
Message-ID: <200909302111.50345.ct529@york.ac.uk>

Dear list,

I have a strange requirement .... I have a vector, for example 

v=[0,0,0,0,1,2,4,6,8,8,8,8]. I have a value,for example x=4.8. 

I would like to understand in which sub interval of v is x. 

In this case, v would be in the sub interval [4,6] that is in the subinterval 
starting from element j=7 to the element j+1=8.

Can we do that with an octave command?

Regards
-- 
Corrado Topi

Global Climate Change & Biodiversity Indicators
Area 18,Department of Biology
University of York, York, YO10 5YW, UK
Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk



From shermanj at umd.edu  Wed Sep 30 15:32:45 2009
From: shermanj at umd.edu (James Sherman Jr.)
Date: Wed, 30 Sep 2009 16:32:45 -0400
Subject: A point in a vector?
In-Reply-To: <200909302111.50345.ct529@york.ac.uk>
References: <200909302111.50345.ct529@york.ac.uk>
Message-ID: <6800c3890909301332j47355e0x6c713f70415f2c0b@mail.gmail.com>

Assuming, that your vector is sorted (I would guess it has to be), you could
do

index = find(v>x,1)

then index would be the index into the vector v.  So, if you wanted the min
and max of the interval, you could just do
j = v(index-1)
j+1 = v(index)

Hope this helps.

On Wed, Sep 30, 2009 at 4:11 PM, Corrado <ct529 at york.ac.uk> wrote:

> Dear list,
>
> I have a strange requirement .... I have a vector, for example
>
> v=[0,0,0,0,1,2,4,6,8,8,8,8]. I have a value,for example x=4.8.
>
> I would like to understand in which sub interval of v is x.
>
> In this case, v would be in the sub interval [4,6] that is in the
> subinterval
> starting from element j=7 to the element j+1=8.
>
> Can we do that with an octave command?
>
> Regards
> --
> Corrado Topi
>
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090930/101c56af/attachment.html 

From storrsjm at email.uc.edu  Wed Sep 30 15:34:31 2009
From: storrsjm at email.uc.edu (Judd Storrs)
Date: Wed, 30 Sep 2009 16:34:31 -0400
Subject: A point in a vector?
In-Reply-To: <200909302111.50345.ct529@york.ac.uk>
References: <200909302111.50345.ct529@york.ac.uk>
Message-ID: <a0daad600909301334x35a972a7h6779d64cd24bd4df@mail.gmail.com>

v=[0,0,0,0,1,2,4,6,8,8,8,8]
j1 = find(v<4.5,1,'last')
j2 = find(v>4.5,1,'first')
[v(j1), v(j2)]

Hope this helps

--judd


On Wed, Sep 30, 2009 at 4:11 PM, Corrado <ct529 at york.ac.uk> wrote:

> Dear list,
>
> I have a strange requirement .... I have a vector, for example
>
> v=[0,0,0,0,1,2,4,6,8,8,8,8]. I have a value,for example x=4.8.
>
> I would like to understand in which sub interval of v is x.
>
> In this case, v would be in the sub interval [4,6] that is in the
> subinterval
> starting from element j=7 to the element j+1=8.
>
> Can we do that with an octave command?
>
> Regards
> --
> Corrado Topi
>
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
>
>
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www-old.cae.wisc.edu/pipermail/help-octave/attachments/20090930/d47eec73/attachment.html 

From martin at mhelm.de  Wed Sep 30 16:22:59 2009
From: martin at mhelm.de (martin_helm)
Date: Wed, 30 Sep 2009 14:22:59 -0700 (PDT)
Subject: A point in a vector?
In-Reply-To: <200909302111.50345.ct529@york.ac.uk>
References: <200909302111.50345.ct529@york.ac.uk>
Message-ID: <25689277.post@talk.nabble.com>




Corrado wrote:
> 
> Dear list,
> 
> I have a strange requirement .... I have a vector, for example 
> 
> v=[0,0,0,0,1,2,4,6,8,8,8,8]. I have a value,for example x=4.8. 
> 
> I would like to understand in which sub interval of v is x. 
> 
> In this case, v would be in the sub interval [4,6] that is in the
> subinterval 
> starting from element j=7 to the element j+1=8.
> 
> Can we do that with an octave command?
> 
> Regards
> -- 
> Corrado Topi
> 
> Global Climate Change & Biodiversity Indicators
> Area 18,Department of Biology
> University of York, York, YO10 5YW, UK
> Phone: + 44 (0) 1904 328645, E-mail: ct529 at york.ac.uk
> 
> 
> _______________________________________________
> Help-octave mailing list
> Help-octave at octave.org
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

You can use the lookup function

v=[0,0,0,0,1,2,4,6,8,8,8,8];
lookup(v, 4.3)

lookup assumes an ordered vector and performs very fast O(log(n))

- mh


-- 
View this message in context: http://www.nabble.com/A-point-in-a-vector--tp25688207p25689277.html
Sent from the Octave - General mailing list archive at Nabble.com.


From sidney at sidney.com  Wed Sep 30 20:40:36 2009
From: sidney at sidney.com (Sidney Markowitz)
Date: Thu, 01 Oct 2009 14:40:36 +1300
Subject: feval call to a subfunction fails when it is in a subfunction
Message-ID: <4AC40894.70707@sidney.com>

I'm new to using Octave and don't know if this is a bug in my code, a 
bug in Octave, or something wrong with my installed copy of Octave.

The following code demonstrates the problem. The first feval, in the 
main function, works and the result is displayed. The other feval, 
inside the subfunction switchyard2, gets an error in line 9
  error: feval: function `mysin' not found

I am using Octave version 3.2.2 compiled on my Macbook under MacOS 
10.6.1, downloaded and built from Macports. I put the following text in 
a file named ./switchyard.m and invoked it in the Octave command line by 
typing   switchyard(10)

Do others get the same results? Is there a bug in my example code? Is it 
a bug in Octave?

function y = switchyard(x)
    z = feval('mysin', x);
    disp(z);
    y = switchyard2(x);
   return;
endfunction

function y = switchyard2(x)
   y = feval('mysin', x);
   return;
endfunction

function y= mysin(x)
   y = sin(x);
  return;
endfunction

From bpabbott at mac.com  Wed Sep 30 20:53:45 2009
From: bpabbott at mac.com (Ben Abbott)
Date: Wed, 30 Sep 2009 21:53:45 -0400
Subject: feval call to a subfunction fails when it is in a subfunction
In-Reply-To: <4AC40894.70707@sidney.com>
References: <4AC40894.70707@sidney.com>
Message-ID: <D5FC8C92-2065-445C-A017-0CF17FE86401@mac.com>


On Sep 30, 2009, at 9:40 PM, Sidney Markowitz wrote:

> I'm new to using Octave and don't know if this is a bug in my code, a
> bug in Octave, or something wrong with my installed copy of Octave.
>
> The following code demonstrates the problem. The first feval, in the
> main function, works and the result is displayed. The other feval,
> inside the subfunction switchyard2, gets an error in line 9
>  error: feval: function `mysin' not found
>
> I am using Octave version 3.2.2 compiled on my Macbook under MacOS
> 10.6.1, downloaded and built from Macports. I put the following text  
> in
> a file named ./switchyard.m and invoked it in the Octave command  
> line by
> typing   switchyard(10)
>
> Do others get the same results? Is there a bug in my example code?  
> Is it
> a bug in Octave?
>
> function y = switchyard(x)
>    z = feval('mysin', x);
>    disp(z);
>    y = switchyard2(x);
>   return;
> endfunction
>
> function y = switchyard2(x)
>   y = feval('mysin', x);
>   return;
> endfunction
>
> function y= mysin(x)
>   y = sin(x);
>  return;
> endfunction


It appears that I'm getting the same.

octave-3.2.2:4> switchyard(pi)
  1.2246e-16
error: feval: function `mysin' not found
error: called from:
error:   /Users/bpabbott/switchyard.m at line 9, column 5
error:   /Users/bpabbott/switchyard.m at line 4, column 6

01 function y = switchyard(x)
02    z = feval('mysin', x);
03    disp(z);
04    y = switchyard2(x);
05   return;
06 endfunction
07
08 function y = switchyard2(x)
09   y = feval('mysin', x);
10   return;
11 endfunction
12
13 function y= mysin(x)
14   y = sin(x);
15  return;
16 endfunction

I'm also running Mac OS 10.6.

I'm presently building the developers sources, so I can't check them  
till tomorrow.

Ben